report_05-07-2025_03-55-17.html

[gw0] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw0] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw0] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1470'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1470', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -11
E assert -11 == 0
E + where -11 = <Process name='Process-1' pid=3537849 parent=3536265 stopped exitcode=-SIGSEGV>.exitcode

test_tidl_unit.py:123: AssertionError[gw0] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.13s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.14s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.280s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.1545s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.1614s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.1709s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.1748s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.1804s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.1842s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.1908s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.1931s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.1955s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.2009s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.2032s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.2052s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.2079s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.2122s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.2145s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.3323s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.3407s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.3466s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.3526s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.3556s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.3595s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.4404s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.4489s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.4532s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.4568s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.4592s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.4622s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.4657s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.4687s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.4714s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.4754s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.4995s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.5040s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.5074s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.5101s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.5137s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.5178s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.5204s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.5236s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.5271s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.5305s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.5310s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.5328s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name ********** Frame Index 1066044866 Running inference - currFrameIdx > numFramesCalibration ********** 0.58560s: VX_ZONE_ERROR: [ownObjectDeInit:171] Is graph use failed, index: 0 0.58569s: VX_ZONE_ERROR: [ownObjectDeInit:177] Is node use failed, index: 0 0.58572s: VX_ZONE_ERROR: [ownObjectDeInit:183] Is kernel use failed, index: 0 0.58574s: VX_ZONE_ERROR: [ownObjectDeInit:184] kernel name: org.khronos.openvx.absdiff 0.58577s: VX_ZONE_ERROR: [ownObjectDeInit:196] Is user data object use failed, index: 0 0.58586s: VX_ZONE_ERROR: [ownObjectDeInit:240] Is tensor use failed, index: 0 0.58590s: VX_ZONE_ERROR: [ownObjectDeInit:282] Is error use failed, index: 0

[gw1] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw1] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw1] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_758'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_758', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -6
E assert -6 == 0
E + where -6 = <Process name='Process-1' pid=3537850 parent=3536268 stopped exitcode=-SIGABRT>.exitcode

test_tidl_unit.py:123: AssertionError[gw1] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------ | Node | Node Name | Reason | ------------------------------------------------------------------------------------ | Unsqueeze | output | Output dimensions after unsqueeze must be less than <= 6 | ------------------------------------------------------------------------------------ ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.10s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.10s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.257s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2323s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2471s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2492s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2513s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2546s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2562s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2593s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2652s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.2672s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.2696s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.2718s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.2741s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.2766s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.3691s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.3777s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.3820s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.3865s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.3900s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.3929s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.3960s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.3983s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.4012s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.4039s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.4070s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.4097s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.4127s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.4150s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.4190s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.4219s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.4247s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.4278s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.4301s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.4324s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.4363s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.4395s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.4416s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.4472s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.4492s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.4521s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.4571s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.4613s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.4615s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.4630s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name ********** Frame Index 1074690309 Running inference - currFrameIdx > numFramesCalibration **********
------------------------------Captured stderr call------------------------------
malloc(): invalid size (unsorted)

[gw8] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw8] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw8] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1612'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1612', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -6
E assert -6 == 0
E + where -6 = <Process name='Process-1' pid=3537859 parent=3536324 stopped exitcode=-SIGABRT>.exitcode

test_tidl_unit.py:123: AssertionError[gw8] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.9s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.10s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.284s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.1316s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.1426s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.1462s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.1509s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.1553s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.1581s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.1651s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.1672s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.1691s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.1715s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.1735s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.1757s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.1779s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.1814s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.1832s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.1860s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.1881s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.1897s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.1916s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.1939s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.1971s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.2006s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.2036s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.2052s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.2071s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.2090s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.2107s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.2126s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.2155s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.2182s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.2209s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.2244s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.2269s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.2303s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.2338s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.2360s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.2401s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.2438s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.2461s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.2512s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.2540s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.2544s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.2555s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name ********** Frame Index 1066723357 Running inference - currFrameIdx > numFramesCalibration **********
------------------------------Captured stderr call------------------------------
malloc(): invalid size (unsorted)

[gw20] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw20] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw20] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1775'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1775', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-1' pid=3537857 parent=3536928 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw20] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.11s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.12s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.267s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.1446s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.1566s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.1594s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.1626s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.1665s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.1691s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.1744s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.1768s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.1803s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.1832s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.1858s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.1890s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.1911s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.2175s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.2218s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.2244s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.2262s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.2291s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.2332s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.2998s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.3074s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.3115s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.3139s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.3165s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.3201s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.3251s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.3283s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.3324s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.3365s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.3395s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.3447s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.3506s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.3539s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.3590s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.3626s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.3662s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.3710s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.3754s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.3791s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.3847s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.3902s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.3906s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.3919s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18699757 bytes MEM: Free's : 26 free's of 18699757 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:55:53.710791069 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -5 is not in valid range [-2,1] Process Process-1: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -5 is not in valid range [-2,1]

[gw5] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw5] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw5] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_723'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_723', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -6
E assert -6 == 0
E + where -6 = <Process name='Process-1' pid=3537852 parent=3536280 stopped exitcode=-SIGABRT>.exitcode

test_tidl_unit.py:123: AssertionError[gw5] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------ | Node | Node Name | Reason | ------------------------------------------------------------------------------------ | Unsqueeze | output | Output dimensions after unsqueeze must be less than <= 6 | ------------------------------------------------------------------------------------ ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.10s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.11s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.284s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.1729s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.1865s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.1897s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.1951s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.1992s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2017s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2046s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2140s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.2171s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.2206s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.2231s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.2252s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.2279s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.2320s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.2347s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.2385s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.2408s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.2443s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.2477s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.2509s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.2594s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.2629s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.2656s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.2696s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.2730s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.2765s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.2792s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.2829s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.2870s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.2914s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.2945s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.2972s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.3002s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.3051s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.3093s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.3132s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.3198s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.3233s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.3257s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.3325s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.3386s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.3392s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.3402s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name
------------------------------Captured stderr call------------------------------
malloc(): invalid size (unsorted)

[gw4] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw4] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw4] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_759'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_759', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -6
E assert -6 == 0
E + where -6 = <Process name='Process-1' pid=3537862 parent=3536277 stopped exitcode=-SIGABRT>.exitcode

test_tidl_unit.py:123: AssertionError[gw4] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------ | Node | Node Name | Reason | ------------------------------------------------------------------------------------ | Unsqueeze | output | Output dimensions after unsqueeze must be less than <= 6 | ------------------------------------------------------------------------------------ ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.14s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.15s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.345s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2966s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.3139s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.3177s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.3229s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.3270s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.3313s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.3344s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.3435s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3482s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3533s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3578s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3603s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.3638s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.3676s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.3712s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.3750s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.3790s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.3821s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.4065s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.4121s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.4160s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.4204s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.4242s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.4265s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.4304s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.4348s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.4378s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.4404s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.4432s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.4469s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.4701s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.4770s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.4818s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.4865s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.4894s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.4928s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.4990s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.5036s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.5081s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.5129s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.5176s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.5178s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.5195s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name ********** Frame Index 1074690309 Running inference - currFrameIdx > numFramesCalibration **********
------------------------------Captured stderr call------------------------------
malloc(): invalid size (unsorted)

[gw2] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw2] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw2] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1533'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1533', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -6
E assert -6 == 0
E + where -6 = <Process name='Process-1' pid=3537861 parent=3536271 stopped exitcode=-SIGABRT>.exitcode

test_tidl_unit.py:123: AssertionError[gw2] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.22s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.25s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.481s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.3191s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.3356s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.3494s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.3560s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.3652s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.3731s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.3856s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.3900s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3959s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.4023s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.4067s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.4134s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.4198s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.4294s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.4360s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.4442s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.4495s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.4553s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.4619s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.4700s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.4773s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.4854s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.4924s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.5006s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.5085s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.5142s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.5205s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.5267s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.5330s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.5415s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.5503s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.5579s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.5684s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.5781s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.5848s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.5940s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.6026s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.6119s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.6205s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.6281s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.6389s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.6401s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.6430s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name
------------------------------Captured stderr call------------------------------
malloc(): corrupted top size

[gw6] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw6] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw6] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_628'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_628', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-1' pid=3537860 parent=3536285 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw6] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------ | Node | Node Name | Reason | ------------------------------------------------------------------------------------ | Unsqueeze | output | Output dimensions after unsqueeze must be less than <= 6 | ------------------------------------------------------------------------------------ ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.13s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.15s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.298s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2251s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2431s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2469s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2491s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2529s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2575s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2607s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2671s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.2727s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.2773s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.2799s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.2836s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.2867s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.3100s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.3148s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.3191s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.3215s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.3251s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.3276s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.3300s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.3330s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.3357s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.3393s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.3414s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.3438s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.3466s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.3496s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.3521s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.3544s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.3966s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.4028s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.4061s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.4112s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.4141s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.4164s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.4196s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.4238s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.4264s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.4302s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.4334s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.4361s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.4367s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.4385s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 19629037 bytes MEM: Free's : 26 free's of 19629037 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:55:57.026596959 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 3 is not in valid range [-2,1] Process Process-1: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 3 is not in valid range [-2,1]

[gw11] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw11] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw11] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1742'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1742', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -6
E assert -6 == 0
E + where -6 = <Process name='Process-1' pid=3537863 parent=3536460 stopped exitcode=-SIGABRT>.exitcode

test_tidl_unit.py:123: AssertionError[gw11] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.22s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.24s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.481s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.5209s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.5366s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.5497s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.5571s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.5663s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.6448s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.6624s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.6713s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.6786s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.6906s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.7000s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.7100s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.7182s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.7293s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.7379s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.7486s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.7580s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.7661s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.7745s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.7855s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.7957s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.8106s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.8188s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.8248s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.8351s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.8446s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.8519s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.8638s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.8745s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.8832s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.8979s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.9045s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.9110s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.9230s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.9318s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.9387s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.9507s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.9584s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.9653s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.9796s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.9911s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.9950s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.9976s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name
------------------------------Captured stderr call------------------------------
malloc_consolidate(): invalid chunk size

[gw10] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw10] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw10] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1757'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1757', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -11
E assert -11 == 0
E + where -11 = <Process name='Process-1' pid=3537865 parent=3536424 stopped exitcode=-SIGSEGV>.exitcode

test_tidl_unit.py:123: AssertionError[gw10] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.25s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.27s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.524s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.5570s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.5789s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.5886s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.6019s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.6130s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.6217s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.6387s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.6480s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.6538s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.6979s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.7109s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.7224s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.7309s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.7435s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.7524s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.7648s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.7735s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.7859s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.7942s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.8063s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.8180s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.8257s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.8317s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.8442s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.8543s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.8633s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.8753s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.8834s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.8911s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.9037s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.9173s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.9243s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.9379s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.9470s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.9540s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.9638s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.9743s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.9818s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.9943s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.10094s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.10200s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.10207s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.10234s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name ********** Frame Index 1050838730 Running inference - currFrameIdx > numFramesCalibration ********** 0.70930s: VX_ZONE_ERROR: [ownObjectDeInit:171] Is graph use failed, index: 0 0.70941s: VX_ZONE_ERROR: [ownObjectDeInit:177] Is node use failed, index: 0 0.70945s: VX_ZONE_ERROR: [ownObjectDeInit:183] Is kernel use failed, index: 0 0.70947s: VX_ZONE_ERROR: [ownObjectDeInit:184] kernel name: org.khronos.openvx.absdiff 0.70950s: VX_ZONE_ERROR: [ownObjectDeInit:196] Is user data object use failed, index: 0 0.70962s: VX_ZONE_ERROR: [ownObjectDeInit:240] Is tensor use failed, index: 0 0.70968s: VX_ZONE_ERROR: [ownObjectDeInit:282] Is error use failed, index: 0

[gw14] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw14] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw14] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1398'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1398', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -11
E assert -11 == 0
E + where -11 = <Process name='Process-1' pid=3537855 parent=3536562 stopped exitcode=-SIGSEGV>.exitcode

test_tidl_unit.py:123: AssertionError[gw14] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.21s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.23s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.544s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.5016s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.5177s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.5360s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.5464s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.5928s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.6033s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.6183s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.6334s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.6411s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.6498s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.6594s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.6674s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.6763s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.6920s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.7023s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.7094s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.7204s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.7296s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.7382s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.7562s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.7635s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.7735s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.7851s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.7949s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.8037s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.8146s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.8210s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.8271s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.8378s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.8461s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.8541s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.8654s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.8730s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.8832s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.9631s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.9834s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.9959s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.10109s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.10231s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.10367s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.10538s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.10551s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.10584s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name ********** Frame Index 1067578608 Running inference - currFrameIdx > numFramesCalibration ********** 0.74300s: VX_ZONE_ERROR: [ownObjectDeInit:171] Is graph use failed, index: 0 0.74313s: VX_ZONE_ERROR: [ownObjectDeInit:177] Is node use failed, index: 0 0.74318s: VX_ZONE_ERROR: [ownObjectDeInit:183] Is kernel use failed, index: 0 0.74321s: VX_ZONE_ERROR: [ownObjectDeInit:184] kernel name: org.khronos.openvx.absdiff 0.74335s: VX_ZONE_ERROR: [ownObjectDeInit:196] Is user data object use failed, index: 0 0.74366s: VX_ZONE_ERROR: [ownObjectDeInit:240] Is tensor use failed, index: 0 0.74380s: VX_ZONE_ERROR: [ownObjectDeInit:282] Is error use failed, index: 0

[gw18] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw18] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw18] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1030'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1030', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-1' pid=3537858 parent=3536765 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw18] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.10s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.10s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.252s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.1434s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.1544s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.1568s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.1593s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.1621s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.1641s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.1694s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.1725s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.1752s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.1767s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.1797s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.1821s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.1844s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.1881s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.1896s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.1927s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.1956s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.1984s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.2010s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.2037s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.2060s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.2078s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.2100s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.2128s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.2152s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.2173s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.2193s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.2217s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.2252s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.2272s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.2295s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.2332s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.2350s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.2372s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.2395s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.2424s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.2450s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.2480s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.2515s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.2544s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.2575s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.2576s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.2588s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18657121 bytes MEM: Free's : 26 free's of 18657121 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:55:48.212692820 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -4 is not in valid range [-2,1] Process Process-1: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -4 is not in valid range [-2,1]

[gw12] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw12] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw12] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1725'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1725', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -11
E assert -11 == 0
E + where -11 = <Process name='Process-1' pid=3537869 parent=3536493 stopped exitcode=-SIGSEGV>.exitcode

test_tidl_unit.py:123: AssertionError[gw12] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.9s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.10s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.276s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.1572s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.1732s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.1762s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.1797s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.1834s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.1875s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.1923s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.1941s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.1964s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.1992s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.2019s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.2048s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.2075s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.2117s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.2148s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.2184s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.2218s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.2249s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.2286s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.2316s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.2349s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.2386s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.2421s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.2447s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.2479s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.2512s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.2543s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.2585s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.2615s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.2659s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.2700s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.2739s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.2775s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.2817s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.2861s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.2898s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.2955s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.2995s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.3029s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.3087s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.3132s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.3140s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.3158s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name ********** Frame Index 1050838730 Running inference - currFrameIdx > numFramesCalibration ********** 0.31883s: VX_ZONE_ERROR: [ownObjectDeInit:171] Is graph use failed, index: 0 0.31894s: VX_ZONE_ERROR: [ownObjectDeInit:177] Is node use failed, index: 0 0.31898s: VX_ZONE_ERROR: [ownObjectDeInit:183] Is kernel use failed, index: 0 0.31901s: VX_ZONE_ERROR: [ownObjectDeInit:184] kernel name: org.khronos.openvx.absdiff 0.31907s: VX_ZONE_ERROR: [ownObjectDeInit:196] Is user data object use failed, index: 0 0.31930s: VX_ZONE_ERROR: [ownObjectDeInit:240] Is tensor use failed, index: 0 0.31941s: VX_ZONE_ERROR: [ownObjectDeInit:282] Is error use failed, index: 0

[gw15] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw15] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw15] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1693'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1693', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-1' pid=3537856 parent=3536629 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw15] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.12s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.13s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.300s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.1727s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.1808s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.1907s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.1970s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2012s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2051s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2142s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2163s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.2185s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.2216s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.2246s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.2264s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.2290s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.2329s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.2351s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.2382s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.2405s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.2448s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.2479s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.2509s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.2542s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.2602s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.2632s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.2653s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.2681s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.2701s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.2724s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.2776s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.2802s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.2850s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.2893s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.2950s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.3000s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.3049s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.3089s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.3130s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.3187s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.3226s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.3271s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.3336s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.3376s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.3379s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.3393s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 23854573 bytes MEM: Free's : 26 free's of 23854573 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:55:47.601481147 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -5 is not in valid range [-2,1] Process Process-1: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -5 is not in valid range [-2,1]

[gw12] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw12] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw12] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_893'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_893', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-2' pid=3539019 parent=3536493 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw12] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------ | Node | Node Name | Reason | ------------------------------------------------------------------------------------ | Unsqueeze | output | Output dimensions after unsqueeze must be less than <= 6 | ------------------------------------------------------------------------------------ ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.24s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.27s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.439s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.3278s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.3511s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.3585s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.3642s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.3737s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.3811s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.3880s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.3994s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.4064s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.4131s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.4229s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.4288s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.4344s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.4463s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.4542s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.4599s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.4682s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.4744s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.4803s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.4878s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.4939s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.5011s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.5092s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.5148s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.5452s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.5559s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.5616s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.5688s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.5768s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.5861s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.5941s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.6044s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.6123s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.6182s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.6286s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.6347s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.6424s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.6518s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.6594s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.6704s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.6805s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.6816s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.6841s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18954349 bytes MEM: Free's : 26 free's of 18954349 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:55:59.075096488 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -6 is not in valid range [-2,1] Process Process-2: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -6 is not in valid range [-2,1]

[gw4] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw4] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw4] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1340'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1340', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -11
E assert -11 == 0
E + where -11 = <Process name='Process-2' pid=3539020 parent=3536277 stopped exitcode=-SIGSEGV>.exitcode

test_tidl_unit.py:123: AssertionError[gw4] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.10s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.11s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.305s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.1770s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.1860s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.1941s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.1980s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2012s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2050s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2177s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2216s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.2251s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.2273s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.2314s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.2340s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.2369s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.2399s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.2427s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.2457s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.2500s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.2520s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.2545s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.2578s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.2620s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.2664s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.2699s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.2727s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.2752s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.2788s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.2820s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.2844s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.2869s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.2905s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.2941s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.3071s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.3122s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.3158s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.3190s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.3217s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.3254s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.3286s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.3322s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.3367s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.3425s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.3430s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.3443s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== 1.325600s: VX_ZONE_ERROR: [ownObjectDeInit:171] Is graph use failed, index: 0 1.325645s: VX_ZONE_ERROR: [ownObjectDeInit:177] Is node use failed, index: 0 1.325648s: VX_ZONE_ERROR: [ownObjectDeInit:183] Is kernel use failed, index: 0 1.325651s: VX_ZONE_ERROR: [ownObjectDeInit:184] kernel name: org.khronos.openvx.absdiff 1.325658s: VX_ZONE_ERROR: [ownObjectDeInit:196] Is user data object use failed, index: 0 1.325681s: VX_ZONE_ERROR: [ownObjectDeInit:240] Is tensor use failed, index: 0 1.325693s: VX_ZONE_ERROR: [ownObjectDeInit:282] Is error use failed, index: 0 1.373161s: VX_ZONE_ERROR: [ownPosixObjectDeInit:300] Deiniting event at index: 0 failed 1.373272s: VX_ZONE_ERROR: [ownPosixObjectDeInit:311] Deiniting queue at index: 41 failed 1.373287s: VX_ZONE_ERROR: [ownPosixObjectDeInit:333] Deiniting mutex at index: 10 failed MEM: Deinit ... !!! MEM: Alloc's: 25 alloc's of 24026861 bytes MEM: Free's : 25 free's of 24026861 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!

[gw8] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw8] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw8] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1686'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1686', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-2' pid=3539169 parent=3536324 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw8] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.16s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.19s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.411s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.3630s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.3774s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.3897s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.3950s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.4020s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.4097s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.4202s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.4250s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.4289s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.4337s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.4377s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.4437s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.4478s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.4541s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.4588s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.4659s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.4718s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.4753s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.4796s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.4826s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.4895s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.4941s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.4989s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.5032s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.5072s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.5122s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.5173s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.5229s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.5279s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.5338s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.5397s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.5471s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.5551s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.5621s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.5680s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.5740s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.5810s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.5893s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.5957s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.6049s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.6115s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.6121s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.6144s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 23854573 bytes MEM: Free's : 26 free's of 23854573 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:01.654071368 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -5 is not in valid range [-2,1] Process Process-2: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -5 is not in valid range [-2,1]

[gw15] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw15] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw15] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1343'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1343', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -6
E assert -6 == 0
E + where -6 = <Process name='Process-2' pid=3539240 parent=3536629 stopped exitcode=-SIGABRT>.exitcode

test_tidl_unit.py:123: AssertionError[gw15] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.15s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.16s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.308s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.1782s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.1893s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.1927s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.1992s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2029s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2055s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2134s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2163s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.2196s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.2237s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.2270s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.2295s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.2327s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.2455s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.2490s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.2518s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.2556s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.2597s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.2621s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.2653s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.2684s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.2727s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.2758s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.2792s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.2851s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.2883s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.2904s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.2948s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.2984s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.3027s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.3067s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.3109s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.3151s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.3180s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.3208s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.3240s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.3288s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.3329s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.3357s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.3392s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.3433s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.3445s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.3462s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ========================
------------------------------Captured stderr call------------------------------
terminate called after throwing an instance of 'onnxruntime::OnnxRuntimeException' what(): /root/onnxruntime/onnxruntime/core/framework/bfc_arena.cc:92 onnxruntime::BFCArena::Chunk* onnxruntime::BFCArena::ChunkFromHandle(onnxruntime::BFCArena::ChunkHandle) h < chunks_.size() was false.

[gw3] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw3] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw3] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_854'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_854', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-2' pid=3539310 parent=3536274 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw3] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------ | Node | Node Name | Reason | ------------------------------------------------------------------------------------ | Unsqueeze | output | Output dimensions after unsqueeze must be less than <= 6 | ------------------------------------------------------------------------------------ ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.14s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.16s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.352s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2914s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.3106s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.3159s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.3210s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.3271s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.3309s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.3350s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.3446s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3493s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3527s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3576s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3656s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.3700s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.4487s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.4620s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.4689s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.4736s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.4806s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.4856s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.4911s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.4969s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.5029s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.5080s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.5157s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.5208s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.5285s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.5348s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.5413s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.5457s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.5511s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.5596s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.5675s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.6065s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.6174s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.6257s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.6330s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.6414s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.6505s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.6573s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.6700s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.6793s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.6799s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.6825s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18681997 bytes MEM: Free's : 26 free's of 18681997 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:06.230243712 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 5 is not in valid range [-2,1] Process Process-2: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 5 is not in valid range [-2,1]

[gw22] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw22] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw22] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1839'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1839', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-2' pid=3539334 parent=3537059 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw22] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.44s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.46s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.554s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.4778s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.5016s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.5194s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.5281s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.5391s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.5495s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.5652s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.5730s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.5804s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.5896s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.5998s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.6087s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.6212s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.6304s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.6381s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.7334s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.7561s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.7647s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.7771s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.7871s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.7970s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.8093s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.8199s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.8294s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.8412s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.8509s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.8574s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.8705s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.8803s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.8880s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.9003s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.9087s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.9170s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.9278s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.9359s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.9446s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.9570s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.9664s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.9739s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.9888s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.9985s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.9992s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.10016s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18681997 bytes MEM: Free's : 26 free's of 18681997 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:03.839183662 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -5 is not in valid range [-2,1] Process Process-2: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -5 is not in valid range [-2,1]

[gw18] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw18] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw18] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1869'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1869', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-2' pid=3539368 parent=3536765 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw18] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.20s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.21s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.554s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.6067s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.6328s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.6417s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.6500s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.6609s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.6696s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.6862s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.6971s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.7060s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.7141s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.7219s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.7322s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.7394s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.7556s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.7655s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.7781s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.7870s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.7961s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.8042s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.8136s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.8238s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.8349s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.8447s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.8555s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.8666s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.8788s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.8872s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.8980s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.9079s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.9171s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.9262s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.9361s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.9493s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.9598s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.9679s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.9796s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.9899s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.10005s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.10133s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.10238s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.10329s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.10338s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.10367s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18954349 bytes MEM: Free's : 26 free's of 18954349 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:06.619381519 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -4 is not in valid range [-2,1] Process Process-2: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -4 is not in valid range [-2,1]

[gw21] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw21] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw21] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1181'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1181', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-2' pid=3539697 parent=3536962 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw21] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.10s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.10s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.294s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.1640s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.1737s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.1783s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.1826s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.1881s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.1918s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.1988s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2012s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.2046s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.2088s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.2121s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.2152s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.2179s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.2211s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.2239s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.2268s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.2290s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.2325s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.2351s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.2386s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.2422s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.2448s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.2476s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.2510s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.2546s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.2572s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.2597s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.2627s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.2660s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.2699s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.2729s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.2753s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.2787s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.2812s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.2834s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.2864s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.2891s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.2917s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.2944s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.2973s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.3004s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.3007s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.3018s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18702445 bytes MEM: Free's : 26 free's of 18702445 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:10.465206857 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -2 is not in valid range [-1,0] Process Process-2: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -2 is not in valid range [-1,0]

[gw2] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw2] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw2] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1606'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1606', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -6
E assert -6 == 0
E + where -6 = <Process name='Process-2' pid=3539794 parent=3536271 stopped exitcode=-SIGABRT>.exitcode

test_tidl_unit.py:123: AssertionError[gw2] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.17s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.19s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.496s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.4347s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.4574s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.4727s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.4813s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.4914s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.4984s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.5121s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.5193s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.5271s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.5336s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.5399s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.5497s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.5563s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.5658s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.5731s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.5823s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.5889s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.5984s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.6062s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.6133s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.6225s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.6327s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.6419s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.6510s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.6593s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.6693s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.6796s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.6866s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.6947s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.7022s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.7114s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.7184s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.7265s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.7340s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.7410s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.7515s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.7614s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.7706s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.7782s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.7900s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.7987s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.7995s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.8017s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name ********** Frame Index 1066723357 Running inference - currFrameIdx > numFramesCalibration **********
------------------------------Captured stderr call------------------------------
malloc(): invalid size (unsorted)

[gw20] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw20] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw20] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1624'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1624', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -6
E assert -6 == 0
E + where -6 = <Process name='Process-2' pid=3539844 parent=3536928 stopped exitcode=-SIGABRT>.exitcode

test_tidl_unit.py:123: AssertionError[gw20] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.14s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.15s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.330s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2142s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2357s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2395s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2430s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2476s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2506s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2583s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2607s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3053s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3096s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3129s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3162s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.3203s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.3263s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.3328s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.3367s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.3398s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.3447s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.3488s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.3517s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.3568s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.3604s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.3637s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.3674s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.3707s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.3765s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.3813s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.3850s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.4400s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.5171s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.5286s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.5371s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.5443s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.5486s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.5519s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.5592s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.5642s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.5688s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.5742s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.5800s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.5850s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.5855s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.5875s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name [TIDL Import] [PARSER] ERROR: Input ONNX tensor element type - -1107064492. Only FLOAT, UINT8, INT32 and INT64 inputs supported for ONNX runtime -- [tidl_onnxRtImport_EP.cpp, 1861] [TIDL Import] [PARSER] ERROR: - Failed in function: TIDL_subgraphImport -- [tidl_onnxRtImport_EP.cpp, 1950] [TIDL Import] [PARSER] ERROR: - Failed in function: TIDL_computeInvokeFunc -- [tidl_onnxRtImport_EP.cpp, 2753]
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:03.200956171 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running TIDL_0 node. Name:'TIDLExecutionProvider_TIDL_0_0' Status Message: TIDL Compute Invoke Failed. munmap_chunk(): invalid pointer

[gw13] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw13] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw13] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1075'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1075', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-2' pid=3539986 parent=3536559 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw13] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.12s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.14s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.418s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.7504s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.7709s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.7772s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.7844s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.7925s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.8006s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.8103s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.8178s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.8239s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.8301s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.8364s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.8438s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.8511s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.8603s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.10690s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.10800s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.10860s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.11807s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.11883s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.11959s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.12028s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.12103s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.12164s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.12231s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.12295s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.12374s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.12438s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.12521s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.12610s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.12663s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.12724s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.12828s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.12883s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.12959s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.13043s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.13102s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.13172s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.13268s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.13327s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.13393s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.13501s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.13508s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.13548s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18656749 bytes MEM: Free's : 26 free's of 18656749 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:08.537258314 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -4 is not in valid range [-2,1] Process Process-2: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -4 is not in valid range [-2,1]

[gw9] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw9] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw9] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1712'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1712', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-2' pid=3540345 parent=3536359 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw9] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.21s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.22s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.467s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.5311s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.5500s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.5635s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.5709s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.5815s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.5900s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.6047s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.6146s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.6226s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.6301s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.6361s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.6443s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.6510s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.6609s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.6676s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.6766s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.6844s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.6910s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.6974s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.7064s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.7139s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.7217s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.7288s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.7375s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.7457s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.7531s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.7596s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.7675s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.7753s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.7832s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.7917s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.8039s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.8131s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.8218s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.8303s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.8387s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.8498s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.8596s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.8671s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.8766s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.8871s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.8878s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.8898s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 23854573 bytes MEM: Free's : 26 free's of 23854573 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:12.139964539 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 3 is not in valid range [-2,1] Process Process-2: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 3 is not in valid range [-2,1]

[gw12] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw12] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw12] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1112'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1112', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-3' pid=3540999 parent=3536493 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw12] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.22s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.23s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.509s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.4702s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.4960s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.5123s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.5206s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.6589s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.6729s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.6866s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.6944s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.7010s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.7136s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.7202s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.7261s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.7354s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.7491s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.7571s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.7673s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.7749s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.7824s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.7929s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.8020s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.8094s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.8201s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.8279s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.8339s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.8431s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.8502s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.8568s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.8672s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.8738s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.8815s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.8933s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.9013s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.9096s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.9222s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.9302s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.9374s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.9515s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.9605s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.9687s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.9829s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.9911s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.9920s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.9946s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18659005 bytes MEM: Free's : 26 free's of 18659005 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:15.002653781 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -3 is not in valid range [-1,0] Process Process-3: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -3 is not in valid range [-1,0]

[gw14] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw14] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw14] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_933'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_933', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-2' pid=3541404 parent=3536562 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw14] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------ | Node | Node Name | Reason | ------------------------------------------------------------------------------------ | Unsqueeze | output | Output dimensions after unsqueeze must be less than <= 6 | ------------------------------------------------------------------------------------ ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.9s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.10s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.229s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2112s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2194s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2280s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2315s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2362s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2404s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2441s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2512s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.2545s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.2577s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.2608s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.2644s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.2675s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.2722s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.2772s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.2809s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.2847s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.2881s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.2908s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.2954s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.2989s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.3031s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.3058s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.3096s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.3134s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.3167s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.3204s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.3232s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.3289s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.3321s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.3360s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.3388s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.3422s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.3454s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.3488s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.3523s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.3561s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.3623s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.3650s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.3684s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.3742s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.3746s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.3754s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 23245933 bytes MEM: Free's : 26 free's of 23245933 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:10.154426967 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -4 is not in valid range [-2,1] Process Process-2: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -4 is not in valid range [-2,1]

[gw19] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw19] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw19] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1042'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1042', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-2' pid=3541729 parent=3536831 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw19] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.16s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.19s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.486s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.3012s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.3220s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.3293s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.3354s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.3440s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.3502s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.3621s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.3682s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3735s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3796s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3848s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3890s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.3955s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.4121s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.4193s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.4254s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.4306s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.4382s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.4437s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.4486s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.4548s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.4614s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.4675s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.4777s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.4860s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.4949s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.5022s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.5093s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.5168s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.5239s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.5320s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.5409s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.5480s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.5584s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.5676s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.5731s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.5821s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.5904s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.5976s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.6047s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.6119s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.6128s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.6157s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18657325 bytes MEM: Free's : 26 free's of 18657325 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:15.447069347 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -3 is not in valid range [-2,1] Process Process-2: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -3 is not in valid range [-2,1]

[gw7] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw7] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw7] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1617'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1617', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -6
E assert -6 == 0
E + where -6 = <Process name='Process-2' pid=3541733 parent=3536289 stopped exitcode=-SIGABRT>.exitcode

test_tidl_unit.py:123: AssertionError[gw7] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.14s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.15s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.476s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.4289s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.4562s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.4630s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.4699s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.4795s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.4861s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.4985s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.5045s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.5130s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.5204s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.5266s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.5326s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.5400s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.5475s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.5537s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.5631s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.5704s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.5784s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.5849s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.5932s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.5999s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.6084s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.6145s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.6209s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.6268s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.6348s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.6410s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.6495s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.6557s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.6600s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.6685s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.6728s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.6771s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.6841s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.6893s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.6942s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.7026s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.7073s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.7120s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.7218s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.7270s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.7274s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.7293s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name
------------------------------Captured stderr call------------------------------
malloc(): invalid size (unsorted)

[gw20] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw20] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw20] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1419'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1419', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -11
E assert -11 == 0
E + where -11 = <Process name='Process-3' pid=3542000 parent=3536928 stopped exitcode=-SIGSEGV>.exitcode

test_tidl_unit.py:123: AssertionError[gw20] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.9s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.10s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.242s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.1224s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.1338s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.1367s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.3868s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.3961s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.3987s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.4050s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.4072s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.4091s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.4117s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.4136s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.4162s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.4194s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.4222s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.4238s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.4262s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.4287s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.7061s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.7162s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.7185s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.7210s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.7241s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.7263s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.7290s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.7307s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.7332s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.7353s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.7367s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.7386s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.7413s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.7443s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.7471s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.7494s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.7516s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.7537s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.7558s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.7585s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.7611s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.7636s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.7663s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.7687s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.7693s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.7708s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name ********** Frame Index 1067578608 Running inference - currFrameIdx > numFramesCalibration ********** 0.63584s: VX_ZONE_ERROR: [ownObjectDeInit:171] Is graph use failed, index: 0 0.63597s: VX_ZONE_ERROR: [ownObjectDeInit:177] Is node use failed, index: 0 0.63602s: VX_ZONE_ERROR: [ownObjectDeInit:183] Is kernel use failed, index: 0 0.63605s: VX_ZONE_ERROR: [ownObjectDeInit:184] kernel name: org.khronos.openvx.absdiff 0.63609s: VX_ZONE_ERROR: [ownObjectDeInit:196] Is user data object use failed, index: 0 0.63632s: VX_ZONE_ERROR: [ownObjectDeInit:240] Is tensor use failed, index: 0 0.63645s: VX_ZONE_ERROR: [ownObjectDeInit:282] Is error use failed, index: 0

[gw18] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw18] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw18] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1907'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1907', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-3' pid=3542376 parent=3536765 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw18] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.12s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.13s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.327s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2445s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2640s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2688s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2727s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2776s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2805s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2921s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2961s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.2998s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3033s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3059s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3092s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.3127s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.3179s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.3205s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.3252s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.3286s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.3324s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.3363s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.3394s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.3442s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.3478s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.3508s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.3546s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.3587s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.3616s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.3641s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.4102s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.4178s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.4212s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.4261s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.4303s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.4347s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.4405s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.4444s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.4484s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.4539s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.4577s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.4614s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.4659s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.4702s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.4707s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.4722s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18954349 bytes MEM: Free's : 26 free's of 18954349 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:22.045278124 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 3 is not in valid range [-2,1] Process Process-3: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 3 is not in valid range [-2,1]

[gw1] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw1] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw1] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1082'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1082', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-4' pid=3542727 parent=3536268 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw1] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.22s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.24s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.573s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.5148s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.5468s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.5561s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.5640s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.5758s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.5826s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.6049s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.6118s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.6164s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.6257s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.6325s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.6386s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.6478s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.6591s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.6657s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.6761s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.6850s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.6924s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.7025s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.7093s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.7151s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.7255s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.7403s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.7486s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.7581s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.7673s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.7729s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.7846s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.7910s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.7982s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.8090s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.8167s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.8243s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.8357s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.8438s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.8509s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.8647s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.8740s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.8827s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.8975s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.9057s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.9061s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.9089s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18656749 bytes MEM: Free's : 26 free's of 18656749 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:25.185656076 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -3 is not in valid range [-2,1] Process Process-4: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -3 is not in valid range [-2,1]

[gw14] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw14] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw14] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1327'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1327', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -11
E assert -11 == 0
E + where -11 = <Process name='Process-3' pid=3543106 parent=3536562 stopped exitcode=-SIGSEGV>.exitcode

test_tidl_unit.py:123: AssertionError[gw14] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.14s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.15s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.332s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.1884s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.1983s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2020s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2054s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2099s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2137s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2227s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2266s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.2292s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.2328s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.2357s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.2386s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.2418s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.2451s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.2482s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.2524s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.2555s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.2579s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.2610s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.2660s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.2704s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.2787s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.2810s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.2843s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.2881s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.2917s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.2950s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.2988s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.3010s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.3032s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.3073s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.3100s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.3126s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.3152s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.3179s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.3213s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.3249s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.3277s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.3299s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.3331s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.3370s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.3375s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.3392s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== 0.811394s: VX_ZONE_ERROR: [ownObjectDeInit:171] Is graph use failed, index: 0 0.811422s: VX_ZONE_ERROR: [ownObjectDeInit:177] Is node use failed, index: 0 0.811425s: VX_ZONE_ERROR: [ownObjectDeInit:183] Is kernel use failed, index: 0 0.811427s: VX_ZONE_ERROR: [ownObjectDeInit:184] kernel name: org.khronos.openvx.absdiff 0.811431s: VX_ZONE_ERROR: [ownObjectDeInit:196] Is user data object use failed, index: 0 0.811443s: VX_ZONE_ERROR: [ownObjectDeInit:240] Is tensor use failed, index: 0 0.811448s: VX_ZONE_ERROR: [ownObjectDeInit:282] Is error use failed, index: 0 0.880855s: VX_ZONE_ERROR: [ownPosixObjectDeInit:300] Deiniting event at index: 0 failed 0.880925s: VX_ZONE_ERROR: [ownPosixObjectDeInit:311] Deiniting queue at index: 41 failed 0.880944s: VX_ZONE_ERROR: [ownPosixObjectDeInit:333] Deiniting mutex at index: 10 failed MEM: Deinit ... !!! MEM: Alloc's: 25 alloc's of 24026861 bytes MEM: Free's : 25 free's of 24026861 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!

[gw21] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw21] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw21] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1333'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1333', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -11
E assert -11 == 0
E + where -11 = <Process name='Process-3' pid=3543223 parent=3536962 stopped exitcode=-SIGSEGV>.exitcode

test_tidl_unit.py:123: AssertionError[gw21] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.10s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.11s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.286s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2725s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2876s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2901s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2936s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2976s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.3003s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.3079s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.3106s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3132s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3182s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3211s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3238s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.3265s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.4201s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.4281s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.4333s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.4370s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.4411s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.4443s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.4490s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.4535s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.4575s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.4602s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.4627s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.4670s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.4711s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.4746s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.4778s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.4807s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.4839s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.4885s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.4910s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.4938s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.4972s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.5004s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.5042s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.5083s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.5132s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.5162s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.5201s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.5238s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.5242s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.5258s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== 0.744525s: VX_ZONE_ERROR: [ownObjectDeInit:171] Is graph use failed, index: 0 0.744559s: VX_ZONE_ERROR: [ownObjectDeInit:177] Is node use failed, index: 0 0.744562s: VX_ZONE_ERROR: [ownObjectDeInit:183] Is kernel use failed, index: 0 0.744565s: VX_ZONE_ERROR: [ownObjectDeInit:184] kernel name: org.khronos.openvx.absdiff 0.744569s: VX_ZONE_ERROR: [ownObjectDeInit:196] Is user data object use failed, index: 0 0.744579s: VX_ZONE_ERROR: [ownObjectDeInit:240] Is tensor use failed, index: 0 0.744586s: VX_ZONE_ERROR: [ownObjectDeInit:282] Is error use failed, index: 0 0.791991s: VX_ZONE_ERROR: [ownPosixObjectDeInit:300] Deiniting event at index: 0 failed 0.792046s: VX_ZONE_ERROR: [ownPosixObjectDeInit:311] Deiniting queue at index: 41 failed 0.792066s: VX_ZONE_ERROR: [ownPosixObjectDeInit:333] Deiniting mutex at index: 10 failed MEM: Deinit ... !!! MEM: Alloc's: 25 alloc's of 24026861 bytes MEM: Free's : 25 free's of 24026861 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!

[gw0] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw0] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw0] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1570'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1570', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -6
E assert -6 == 0
E + where -6 = <Process name='Process-3' pid=3543229 parent=3536265 stopped exitcode=-SIGABRT>.exitcode

test_tidl_unit.py:123: AssertionError[gw0] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.13s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.14s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.290s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2607s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2704s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2783s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2813s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2845s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2870s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2927s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2964s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.2991s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3016s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3041s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3074s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.3103s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.3138s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.3159s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.3188s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.3215s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.3482s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.3521s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.3544s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.3561s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.3579s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.3599s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.3614s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.3634s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.3651s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.3670s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.3686s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.3710s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.3728s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.3750s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.3768s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.3785s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.3810s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.3826s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.3842s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.3870s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.3890s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.3909s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.3946s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.3976s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.3977s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.3984s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name
------------------------------Captured stderr call------------------------------
malloc(): corrupted top size

[gw9] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw9] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw9] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_863'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_863', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-3' pid=3543451 parent=3536359 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw9] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------ | Node | Node Name | Reason | ------------------------------------------------------------------------------------ | Unsqueeze | output | Output dimensions after unsqueeze must be less than <= 6 | ------------------------------------------------------------------------------------ ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.14s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.15s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.305s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2534s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2673s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2721s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2745s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2792s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2816s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2848s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2952s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.2988s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3015s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3052s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3091s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.3123s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.3715s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.3785s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.3840s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.3895s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.3936s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.3965s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.4001s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.4031s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.4071s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.4102s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.4152s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.4195s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.4234s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.4268s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.4296s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.4335s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.4370s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.4411s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.4475s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.4525s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.4565s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.4612s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.4658s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.4703s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.4776s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.4842s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.4901s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.4967s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.4972s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.4990s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18681997 bytes MEM: Free's : 26 free's of 18681997 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:29.394600769 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 2 is not in valid range [-2,1] Process Process-3: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 2 is not in valid range [-2,1]

[gw8] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw8] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw8] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_871'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_871', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-4' pid=3543651 parent=3536324 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw8] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------ | Node | Node Name | Reason | ------------------------------------------------------------------------------------ | Unsqueeze | output | Output dimensions after unsqueeze must be less than <= 6 | ------------------------------------------------------------------------------------ ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.15s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.16s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.332s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2730s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2949s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.3001s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.3036s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.3106s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.3164s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.3238s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.3326s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3356s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3401s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3432s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3471s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.3509s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.4721s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.4847s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.4927s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.4981s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.5023s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.5065s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.5101s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.5167s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.5225s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.5284s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.5310s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.5356s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.5399s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.5439s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.5495s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.5548s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.5586s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.5640s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.5702s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.5767s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.5820s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.5864s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.5895s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.5944s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.5994s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.6034s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.6087s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.6149s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.6154s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.6170s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18954349 bytes MEM: Free's : 26 free's of 18954349 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:24.152947894 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -7 is not in valid range [-2,1] Process Process-4: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -7 is not in valid range [-2,1]

[gw12] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw12] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw12] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1411'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1411', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -11
E assert -11 == 0
E + where -11 = <Process name='Process-4' pid=3544159 parent=3536493 stopped exitcode=-SIGSEGV>.exitcode

test_tidl_unit.py:123: AssertionError[gw12] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.59s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.61s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.519s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.4694s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.4885s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.4972s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.5065s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.5164s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.5245s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.5348s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.5409s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.5480s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.5549s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.5613s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.5697s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.5780s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.5867s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.5935s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.6007s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.6078s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.6157s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.6280s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.6347s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.6415s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.6528s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.6598s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.6852s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.6974s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.7070s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.7160s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.7272s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.7338s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.7426s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.7548s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.7630s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.7710s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.7813s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.7893s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.7966s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.8086s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.8187s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.8248s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.8370s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.8443s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.8447s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.8474s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name ********** Frame Index 1062831271 Running inference - currFrameIdx > numFramesCalibration ********** 0.100354s: VX_ZONE_ERROR: [ownObjectDeInit:171] Is graph use failed, index: 0 0.100381s: VX_ZONE_ERROR: [ownObjectDeInit:177] Is node use failed, index: 0 0.100385s: VX_ZONE_ERROR: [ownObjectDeInit:183] Is kernel use failed, index: 0 0.100388s: VX_ZONE_ERROR: [ownObjectDeInit:184] kernel name: org.khronos.openvx.absdiff 0.100393s: VX_ZONE_ERROR: [ownObjectDeInit:196] Is user data object use failed, index: 0 0.100418s: VX_ZONE_ERROR: [ownObjectDeInit:240] Is tensor use failed, index: 0 0.100430s: VX_ZONE_ERROR: [ownObjectDeInit:282] Is error use failed, index: 0

[gw23] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw23] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw23] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1642'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1642', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-3' pid=3544391 parent=3537186 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw23] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.9s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.10s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.327s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.1750s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.1880s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.1982s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2027s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2116s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2143s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2235s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2281s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.2309s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.2351s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.2367s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.2389s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.2416s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.2454s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.2472s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.2494s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.2517s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.2544s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.2569s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.2592s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.2618s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.3318s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.3377s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.3420s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.3450s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.3470s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.3506s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.3540s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.3574s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.3609s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.3651s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.3671s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.3698s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.3740s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.3765s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.3813s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.3848s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.3871s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.3898s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.3943s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.3973s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.3979s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.3994s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 19629037 bytes MEM: Free's : 26 free's of 19629037 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:26.821662133 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -6 is not in valid range [-2,1] Process Process-3: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -6 is not in valid range [-2,1]

[gw11] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw11] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw11] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_783'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_783', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-3' pid=3544425 parent=3536460 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw11] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------ | Node | Node Name | Reason | ------------------------------------------------------------------------------------ | Unsqueeze | output | Output dimensions after unsqueeze must be less than <= 6 | ------------------------------------------------------------------------------------ ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.13s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.14s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.296s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2552s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2734s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2779s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2813s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2854s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2916s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2952s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.3045s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3071s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3111s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3143s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3183s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.3210s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.3259s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.3287s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.3321s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.3357s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.3384s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.3409s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.3452s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.3477s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.3508s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.3544s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.3565s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.3594s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.3655s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.3690s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.3722s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.3771s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.3802s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.3836s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.3869s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.3897s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.3941s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.3988s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.4018s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.4058s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.4095s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.4129s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.4171s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.4218s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.4222s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.4236s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18699757 bytes MEM: Free's : 26 free's of 18699757 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:34.700843407 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -6 is not in valid range [-2,1] Process Process-3: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -6 is not in valid range [-2,1]

[gw19] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw19] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw19] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1117'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1117', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-3' pid=3544519 parent=3536831 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw19] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.12s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.13s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.348s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2445s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2603s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2654s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2685s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2739s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2772s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2933s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2975s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3002s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3044s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3073s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3094s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.3390s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.3454s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.3481s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.3503s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.3531s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.3578s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.3609s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.3638s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.3673s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.3703s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.3735s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.3774s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.3798s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.3855s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.3939s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.4001s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.4030s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.4993s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.5084s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.5153s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.5191s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.5234s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.5279s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.5310s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.5358s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.5402s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.5458s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.5495s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.6980s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.7021s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.7042s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18659005 bytes MEM: Free's : 26 free's of 18659005 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:28.319289347 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 2 is not in valid range [-2,1] Process Process-3: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 2 is not in valid range [-2,1]

[gw20] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw20] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw20] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1751'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1751', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -6
E assert -6 == 0
E + where -6 = <Process name='Process-4' pid=3544547 parent=3536928 stopped exitcode=-SIGABRT>.exitcode

test_tidl_unit.py:123: AssertionError[gw20] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.19s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.21s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.498s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.4097s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.4272s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.4361s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.4442s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.4525s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.4599s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.4937s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.5002s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.5077s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.5149s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.5208s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.5304s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.5370s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.5474s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.5558s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.5631s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.5699s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.5831s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.5921s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.6009s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.6194s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.6307s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.6386s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.6510s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.6593s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.6700s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.6831s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.6917s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.6987s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.7107s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.7234s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.7358s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.7461s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.7555s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.7618s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.7722s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.7804s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.7883s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.8059s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.8157s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.8235s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.8259s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.8281s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name
------------------------------Captured stderr call------------------------------
malloc_consolidate(): invalid chunk size

[gw15] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw15] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw15] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1750'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1750', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -11
E assert -11 == 0
E + where -11 = <Process name='Process-4' pid=3544553 parent=3536629 stopped exitcode=-SIGSEGV>.exitcode

test_tidl_unit.py:123: AssertionError[gw15] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.15s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.16s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.350s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2133s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2305s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2341s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2404s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2456s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2490s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2581s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2603s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3690s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3801s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3838s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3879s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.3916s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.5696s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.5830s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.5872s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.5905s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.5965s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.5998s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.6045s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.6083s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.6151s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.6185s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.6215s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.6244s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.6271s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.6306s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.6353s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.6391s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.6445s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.6486s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.6528s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.6572s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.6604s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.6632s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.6674s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.6728s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.6779s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.6818s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.6886s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.6945s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.6953s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.6973s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name ********** Frame Index 1050838730 Running inference - currFrameIdx > numFramesCalibration ********** 0.37884s: VX_ZONE_ERROR: [ownObjectDeInit:171] Is graph use failed, index: 0 0.37893s: VX_ZONE_ERROR: [ownObjectDeInit:177] Is node use failed, index: 0 0.37897s: VX_ZONE_ERROR: [ownObjectDeInit:183] Is kernel use failed, index: 0 0.37899s: VX_ZONE_ERROR: [ownObjectDeInit:184] kernel name: org.khronos.openvx.absdiff 0.37903s: VX_ZONE_ERROR: [ownObjectDeInit:196] Is user data object use failed, index: 0 0.37916s: VX_ZONE_ERROR: [ownObjectDeInit:240] Is tensor use failed, index: 0 0.37924s: VX_ZONE_ERROR: [ownObjectDeInit:282] Is error use failed, index: 0

[gw7] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw7] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw7] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1913'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1913', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-3' pid=3544662 parent=3536289 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw7] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.17s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.18s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.490s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.4150s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.4396s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.4461s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.4527s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.4608s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.4674s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.4847s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.4901s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.4959s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.5052s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.5103s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.5163s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.5261s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.7039s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.7202s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.7341s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.7424s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.7513s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.7627s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.7717s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.7801s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.7928s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.7986s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.8038s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.8135s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.8200s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.8241s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.9947s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.10160s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.10228s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.10345s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.10396s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.10449s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.10531s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.10608s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.10666s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.10771s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.10824s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.10882s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.10988s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.11057s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.11062s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.11096s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 23245933 bytes MEM: Free's : 26 free's of 23245933 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:29.361804596 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -6 is not in valid range [-2,1] Process Process-3: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -6 is not in valid range [-2,1]

[gw4] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw4] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw4] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1649'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1649', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-4' pid=3544693 parent=3536277 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw4] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.16s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.19s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.400s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.3580s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.4244s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.4344s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.4551s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.4687s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.4801s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.4970s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.5067s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.5153s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.5240s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.5345s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.5730s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.5881s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.6004s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.6136s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.6257s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.6593s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.6813s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.6937s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.7111s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.7332s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.7541s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.7728s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.7887s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.7993s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.8116s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.8272s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.8376s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.8459s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.8613s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.8753s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.8877s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.9047s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.9174s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.9281s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.9428s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.9571s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.9678s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.9819s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.9935s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.10068s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.10076s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.10115s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 19629037 bytes MEM: Free's : 26 free's of 19629037 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:32.490779934 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -4 is not in valid range [-2,1] Process Process-4: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -4 is not in valid range [-2,1]

[gw3] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw3] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw3] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1303'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1303', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-4' pid=3545029 parent=3536274 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw3] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.12s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.13s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.317s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2362s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2465s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2496s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2536s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2580s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2609s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2656s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2691s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.2734s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.2761s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.2805s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.2838s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.2864s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.3512s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.3583s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.3609s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.3632s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.3680s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.3710s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.3738s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.3777s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.3837s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.3866s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.3915s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.3954s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.3996s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.4031s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.4080s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.4119s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.4162s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.4201s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.4256s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.4321s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.4378s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.4438s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.4489s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.4550s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.4612s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.4658s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.4720s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.4813s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.4833s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.4853s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 20266093 bytes MEM: Free's : 26 free's of 20266093 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:37.964365764 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -5 is not in valid range [-2,1] Process Process-4: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -5 is not in valid range [-2,1]

[gw2] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw2] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw2] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1367'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1367', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -11
E assert -11 == 0
E + where -11 = <Process name='Process-4' pid=3545172 parent=3536271 stopped exitcode=-SIGSEGV>.exitcode

test_tidl_unit.py:123: AssertionError[gw2] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.11s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.12s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.329s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2923s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.3141s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.3180s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.3222s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.3270s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.3305s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.3404s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.3432s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3462s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3506s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3538s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3571s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.3609s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.4385s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.4489s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.4549s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.4585s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.4638s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.4684s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.4721s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.4774s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.4829s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.4882s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.4933s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.4972s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.5005s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.5042s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.5072s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.5109s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.5147s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.5190s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.5226s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.5257s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.5289s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.5318s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.5374s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.5428s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.5473s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.5514s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.5583s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.5632s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.5636s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.5654s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name ********** Frame Index 1050100735 Running inference - currFrameIdx > numFramesCalibration ********** 0.33633s: VX_ZONE_ERROR: [ownObjectDeInit:171] Is graph use failed, index: 0 0.33639s: VX_ZONE_ERROR: [ownObjectDeInit:177] Is node use failed, index: 0 0.33641s: VX_ZONE_ERROR: [ownObjectDeInit:183] Is kernel use failed, index: 0 0.33643s: VX_ZONE_ERROR: [ownObjectDeInit:184] kernel name: org.khronos.openvx.absdiff 0.33647s: VX_ZONE_ERROR: [ownObjectDeInit:196] Is user data object use failed, index: 0 0.33658s: VX_ZONE_ERROR: [ownObjectDeInit:240] Is tensor use failed, index: 0 0.33664s: VX_ZONE_ERROR: [ownObjectDeInit:282] Is error use failed, index: 0 0.81429s: VX_ZONE_ERROR: [ownPosixObjectDeInit:300] Deiniting event at index: 0 failed 0.81516s: VX_ZONE_ERROR: [ownPosixObjectDeInit:311] Deiniting queue at index: 41 failed 0.81528s: VX_ZONE_ERROR: [ownPosixObjectDeInit:333] Deiniting mutex at index: 10 failed MEM: Deinit ... !!! MEM: Alloc's: 25 alloc's of 24899309 bytes MEM: Free's : 25 free's of 24899309 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!

[gw0] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw0] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw0] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1645'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1645', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-4' pid=3545420 parent=3536265 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw0] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.12s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.13s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.325s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2141s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2279s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2316s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2349s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2392s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2428s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2506s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2541s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.2569s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.2618s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.2646s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.2675s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.2711s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.2756s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.2789s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.2839s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.2875s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.2920s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.2969s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.3006s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.3056s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.3088s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.3115s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.3145s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.3175s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.3209s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.3247s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.3281s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.3311s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.3347s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.3412s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.3448s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.3483s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.3537s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.3566s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.3613s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.3672s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.3707s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.3735s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.3777s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.3836s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.3838s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.3850s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 19629037 bytes MEM: Free's : 26 free's of 19629037 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:37.646442775 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -6 is not in valid range [-2,1] Process Process-4: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -6 is not in valid range [-2,1]

[gw6] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw6] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw6] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1666'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1666', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-4' pid=3545476 parent=3536285 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw6] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.13s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.14s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.294s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.3387s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.3582s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.3622s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.3651s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.3689s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.3725s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.3799s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.3824s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3861s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3897s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3925s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3958s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.3980s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.4868s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.4944s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.5003s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.5031s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.5067s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.5099s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.5144s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.5169s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.5200s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.5240s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.5265s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.5296s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.5340s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.5373s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.5396s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.5424s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.5452s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.5485s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.5514s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.5561s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.5592s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.5631s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.5668s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.5699s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.5727s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.5756s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.5785s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.5825s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.5831s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.5848s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 19629037 bytes MEM: Free's : 26 free's of 19629037 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:39.978387939 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 5 is not in valid range [-2,1] Process Process-4: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 5 is not in valid range [-2,1]

[gw22] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw22] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw22] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1608'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1608', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -6
E assert -6 == 0
E + where -6 = <Process name='Process-4' pid=3545908 parent=3537059 stopped exitcode=-SIGABRT>.exitcode

test_tidl_unit.py:123: AssertionError[gw22] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.19s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.21s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.509s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.4538s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.4859s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.4968s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.5045s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.5143s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.5241s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.5390s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.5470s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.5544s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.5630s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.5713s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.5831s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.5915s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.6828s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.6997s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.7105s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.7196s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.7295s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.7386s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.7481s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.7604s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.7698s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.7765s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.8468s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.8643s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.8741s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.8845s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.8932s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.9014s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.9119s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.9215s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.9304s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.9403s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.9470s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.9531s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.9680s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.9767s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.9849s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.9944s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.10053s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.10165s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.10172s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.10204s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name ********** Frame Index 1066723357 Running inference - currFrameIdx > numFramesCalibration **********
------------------------------Captured stderr call------------------------------
malloc(): invalid size (unsorted)

[gw1] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw1] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw1] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1254'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1254', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-5' pid=3546211 parent=3536268 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw1] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.9s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.10s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.298s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2259s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2396s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2517s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2564s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2599s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2622s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2695s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2713s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.2737s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.2763s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.2796s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.2822s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.2849s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.2891s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.2925s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.2945s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.2985s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.3021s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.3042s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.3079s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.3106s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.3140s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.3173s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.3211s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.3240s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.3272s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.3300s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.3336s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.3383s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.3418s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.3461s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.3498s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.3523s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.3560s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.3621s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.3647s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.3692s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.3725s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.3750s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.3810s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.3861s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.3864s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.3879s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 19374189 bytes MEM: Free's : 26 free's of 19374189 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:36.725765824 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 3 is not in valid range [-2,1] Process Process-5: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 3 is not in valid range [-2,1]

[gw14] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw14] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw14] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1938'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1938', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-4' pid=3546508 parent=3536562 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw14] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.9s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.10s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.194s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.1087s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.1152s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.1180s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.1203s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.1233s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.1253s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.1295s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.1309s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.1325s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.1343s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.1357s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.1375s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.1388s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.1412s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.1431s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.1449s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.1466s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.1482s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.1499s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.1516s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.1539s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.1558s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.1577s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.1593s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.1612s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.1645s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.1665s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.2195s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.2218s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.2229s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.2249s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.2264s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.2279s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.2293s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.2308s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.2330s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.2350s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.2369s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.2389s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.2411s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.2438s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.2440s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.2449s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 23245933 bytes MEM: Free's : 26 free's of 23245933 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:35.882730097 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 5 is not in valid range [-2,1] Process Process-4: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 5 is not in valid range [-2,1]

[gw21] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw21] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw21] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_912'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_912', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-4' pid=3546564 parent=3536962 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw21] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------ | Node | Node Name | Reason | ------------------------------------------------------------------------------------ | Unsqueeze | output | Output dimensions after unsqueeze must be less than <= 6 | ------------------------------------------------------------------------------------ ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.21s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.23s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.514s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.5422s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.5653s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.5741s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.5834s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.5953s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.6040s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.6136s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.6235s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.6298s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.6396s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.6467s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.6546s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.6656s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.6745s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.6807s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.6930s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.7018s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.7110s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.7217s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.7325s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.7420s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.7550s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.7637s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.7727s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.7854s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.7943s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.8044s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.8221s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.8318s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.8421s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.8567s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.8654s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.8727s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.8871s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.8964s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.9071s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.9198s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.9293s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.9366s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.9513s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.9629s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.9641s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.9663s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 23245933 bytes MEM: Free's : 26 free's of 23245933 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:43.726317021 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: 'axes' has a duplicate axis Process Process-4: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.InvalidArgument: [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: 'axes' has a duplicate axis

[gw23] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw23] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw23] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_944'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_944', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-4' pid=3546634 parent=3537186 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw23] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------ | Node | Node Name | Reason | ------------------------------------------------------------------------------------ | Unsqueeze | output | Output dimensions after unsqueeze must be less than <= 6 | ------------------------------------------------------------------------------------ ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.12s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.13s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.299s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2526s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2664s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2752s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2804s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2851s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2885s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2933s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.3022s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3837s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3929s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3961s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3998s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.4039s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.4084s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.4114s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.4148s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.4186s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.4235s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.4269s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.4319s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.4369s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.4427s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.4463s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.4520s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.4558s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.4599s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.4638s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.4699s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.4744s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.4775s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.4824s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.4858s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.4892s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.4933s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.4967s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.5013s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.5055s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.5092s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.5128s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.5183s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.5237s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.5240s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.5257s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 23245933 bytes MEM: Free's : 26 free's of 23245933 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:42.677086135 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 5 is not in valid range [-2,1] Process Process-4: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 5 is not in valid range [-2,1]

[gw7] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw7] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw7] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1308'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1308', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -6
E assert -6 == 0
E + where -6 = <Process name='Process-4' pid=3546954 parent=3536289 stopped exitcode=-SIGABRT>.exitcode

test_tidl_unit.py:123: AssertionError[gw7] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.13s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.14s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.365s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2933s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.3120s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.3146s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.3188s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.3262s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.3289s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.3391s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.3420s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3456s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3498s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3524s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3572s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.3624s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.4080s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.4158s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.4217s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.4250s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.4317s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.4349s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.4384s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.4436s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.4492s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.4531s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.4569s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.4601s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.4643s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.4680s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.4711s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.4757s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.4804s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.4853s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.4886s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.4930s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.4957s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.5008s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.5051s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.5098s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.5129s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.5168s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.5226s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.5271s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.5279s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.5294s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ========================
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:44.211798029 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -5 is not in valid range [-2,1] terminate called after throwing an instance of 'onnxruntime::OnnxRuntimeException' what(): /root/onnxruntime/onnxruntime/core/framework/bfc_arena.cc:92 onnxruntime::BFCArena::Chunk* onnxruntime::BFCArena::ChunkFromHandle(onnxruntime::BFCArena::ChunkHandle) h < chunks_.size() was false.

[gw9] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw9] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw9] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1151'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1151', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-4' pid=3546955 parent=3536359 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw9] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.9s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.10s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.312s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.1812s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.1895s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.1926s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.1950s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.1999s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2026s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2107s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2153s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.2181s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.2209s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.2239s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.2271s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.2295s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.2341s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.2368s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.2400s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.2426s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.2456s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.2482s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.2562s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.2599s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.2637s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.2659s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.2691s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.2727s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.2749s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.2771s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.2792s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.2816s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.2846s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.2890s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.2932s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.2974s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.3076s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.3136s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.3164s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.3200s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.3235s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.3278s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.3317s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.3379s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.3383s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.3397s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18674797 bytes MEM: Free's : 26 free's of 18674797 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:38.934584590 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -2 is not in valid range [-1,0] Process Process-4: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -2 is not in valid range [-1,0]

[gw2] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw2] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw2] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1072'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1072', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-5' pid=3547390 parent=3536271 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw2] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.17s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.18s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.485s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.4415s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.4623s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.4774s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.4875s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.4981s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.5063s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.5202s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.5285s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.5358s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.5428s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.5496s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.5596s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.5676s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.5790s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.5864s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.5960s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.6042s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.6144s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.6228s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.6320s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.6432s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.6544s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.6653s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.6788s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.6879s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.6972s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.7056s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.7134s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.7198s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.7263s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.7336s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.7398s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.7464s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.7534s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.7608s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.7700s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.7782s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.7854s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.7935s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.7998s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.8064s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.8070s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.8090s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18656749 bytes MEM: Free's : 26 free's of 18656749 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:52.384974618 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -4 is not in valid range [-2,1] Process Process-5: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -4 is not in valid range [-2,1]

[gw8] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw8] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw8] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1565'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1565', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -6
E assert -6 == 0
E + where -6 = <Process name='Process-6' pid=3547830 parent=3536324 stopped exitcode=-SIGABRT>.exitcode

test_tidl_unit.py:123: AssertionError[gw8] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.15s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.17s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.471s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.4232s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.4426s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.4550s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.4636s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.4741s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.4811s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.4938s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.5012s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.5074s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.5148s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.5211s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.5285s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.5351s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.5455s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.5580s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.5686s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.5749s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.5811s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.5879s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.5949s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.6030s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.6108s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.6183s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.6295s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.6360s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.6437s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.6553s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.6654s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.6735s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.6852s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.6961s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.7078s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.7207s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.7289s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.7356s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.7472s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.7584s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.7687s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.7780s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.7887s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.7951s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.7958s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.7977s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name
------------------------------Captured stderr call------------------------------
malloc(): corrupted top size

[gw11] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw11] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw11] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1274'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1274', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-4' pid=3547858 parent=3536460 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw11] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.20s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.21s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.574s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.5166s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.5320s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.5453s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.5521s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.5655s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.5743s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.5881s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.5955s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.6032s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.6135s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.6194s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.6271s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.6337s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.6422s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.6486s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.6569s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.6657s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.6733s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.6802s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.6880s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.6953s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.7036s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.7111s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.7197s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.7271s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.7399s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.7472s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.7555s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.7650s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.7737s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.7835s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.7917s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.8034s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.8135s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.8221s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.8725s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.8869s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.8953s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.9025s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.9138s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.9284s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.9299s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.9319s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 20266093 bytes MEM: Free's : 26 free's of 20266093 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:58.220536595 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -5 is not in valid range [-2,1] Process Process-4: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -5 is not in valid range [-2,1]

[gw14] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw14] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw14] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1733'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1733', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -11
E assert -11 == 0
E + where -11 = <Process name='Process-5' pid=3548173 parent=3536562 stopped exitcode=-SIGSEGV>.exitcode

test_tidl_unit.py:123: AssertionError[gw14] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.11s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.12s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.304s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2821s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2958s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2998s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.3024s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.3066s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.3096s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.3140s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.3173s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3198s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3238s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3894s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3971s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.4003s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.4074s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.4112s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.4165s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.4212s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.4243s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.4272s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.4318s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.4370s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.4425s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.4467s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.4524s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.4572s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.4616s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.4663s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.4697s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.4742s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.4773s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.4817s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.5687s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.5793s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.5821s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.5862s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.5910s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.5939s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.5981s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.6011s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.6042s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.6094s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.6097s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.6115s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name ********** Frame Index 1050838730 Running inference - currFrameIdx > numFramesCalibration ********** 0.36904s: VX_ZONE_ERROR: [ownObjectDeInit:171] Is graph use failed, index: 0 0.36909s: VX_ZONE_ERROR: [ownObjectDeInit:177] Is node use failed, index: 0 0.36911s: VX_ZONE_ERROR: [ownObjectDeInit:183] Is kernel use failed, index: 0 0.36913s: VX_ZONE_ERROR: [ownObjectDeInit:184] kernel name: org.khronos.openvx.absdiff 0.36915s: VX_ZONE_ERROR: [ownObjectDeInit:196] Is user data object use failed, index: 0 0.36923s: VX_ZONE_ERROR: [ownObjectDeInit:240] Is tensor use failed, index: 0 0.36929s: VX_ZONE_ERROR: [ownObjectDeInit:282] Is error use failed, index: 0

[gw0] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw0] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw0] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1337'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1337', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -6
E assert -6 == 0
E + where -6 = <Process name='Process-5' pid=3548663 parent=3536265 stopped exitcode=-SIGABRT>.exitcode

test_tidl_unit.py:123: AssertionError[gw0] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.20s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.22s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.482s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.4345s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.4500s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.4569s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.4637s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.4722s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.4799s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.4907s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.4959s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.5013s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.5086s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.5139s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.5199s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.5275s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.5369s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.5429s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.5508s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.5574s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.5653s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.5759s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.5824s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.5911s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.6028s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.6086s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.6156s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.6271s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.6356s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.6416s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.6509s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.6589s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.6643s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.6747s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.6828s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.6894s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.7013s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.7091s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.7179s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.7293s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.7362s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.7418s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.7532s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.7627s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.7634s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.7661s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ========================
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:53.746641358 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -5 is not in valid range [-1,0] terminate called after throwing an instance of 'onnxruntime::OnnxRuntimeException' what(): /root/onnxruntime/onnxruntime/core/framework/bfc_arena.cc:92 onnxruntime::BFCArena::Chunk* onnxruntime::BFCArena::ChunkFromHandle(onnxruntime::BFCArena::ChunkHandle) h < chunks_.size() was false.

[gw3] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw3] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw3] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_826'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_826', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-5' pid=3548728 parent=3536274 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw3] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------ | Node | Node Name | Reason | ------------------------------------------------------------------------------------ | Unsqueeze | output | Output dimensions after unsqueeze must be less than <= 6 | ------------------------------------------------------------------------------------ ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.23s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.25s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.554s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.4403s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.4546s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.4607s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.4684s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.4784s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.4849s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.4923s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.5027s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.5095s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.5909s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.6022s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.6096s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.6164s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.7111s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.7264s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.7344s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.7420s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.7516s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.7585s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.7655s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.7730s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.7811s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.7876s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.7962s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.8032s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.8123s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.8209s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.8331s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.8395s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.8449s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.8554s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.8643s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.8709s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.8824s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.8900s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.8969s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.9076s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.9148s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.9207s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.9329s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.9410s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.9417s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.9442s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18681997 bytes MEM: Free's : 26 free's of 18681997 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:50.157082949 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -6 is not in valid range [-2,1] Process Process-5: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -6 is not in valid range [-2,1]

[gw17] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw17] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw17] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1226'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1226', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-4' pid=3548980 parent=3536730 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw17] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.17s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.21s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.426s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.4044s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.4247s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.4356s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.4428s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.4499s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.4562s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.4680s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.4737s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.4794s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.4868s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.4932s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.4997s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.5054s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.5145s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.5216s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.5360s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.5440s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.5502s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.5561s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.5634s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.5696s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.5781s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.5836s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.5910s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.5986s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.6063s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.6127s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.6205s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.6321s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.6412s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.6513s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.6615s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.6691s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.6769s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.6879s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.6957s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.7032s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.7148s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.7223s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.7301s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.7408s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.7415s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.7438s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 19374189 bytes MEM: Free's : 26 free's of 19374189 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:52.432870591 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -4 is not in valid range [-2,1] Process Process-4: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -4 is not in valid range [-2,1]

[gw6] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw6] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw6] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_958'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_958', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-5' pid=3549273 parent=3536285 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw6] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------ | Node | Node Name | Reason | ------------------------------------------------------------------------------------ | Unsqueeze | output | Output dimensions after unsqueeze must be less than <= 6 | ------------------------------------------------------------------------------------ ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.19s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.21s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.442s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.3025s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.3250s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.3308s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.3356s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.3417s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.3500s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.3563s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.3671s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3767s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3818s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3870s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3940s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.3991s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.4161s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.4233s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.4684s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.4769s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.4855s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.4914s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.4975s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.5035s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.5114s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.5210s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.5290s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.5362s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.5441s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.5510s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.5566s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.5627s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.5712s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.5783s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.5845s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.5944s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.6013s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.6080s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.6158s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.6246s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.6299s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.6373s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.6439s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.6515s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.6524s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.6552s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 23245933 bytes MEM: Free's : 26 free's of 23245933 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:56.373692593 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -4 is not in valid range [-2,1] Process Process-5: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -4 is not in valid range [-2,1]

[gw16] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw16] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw16] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_708'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_708', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-4' pid=3549397 parent=3536664 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw16] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------ | Node | Node Name | Reason | ------------------------------------------------------------------------------------ | Unsqueeze | output | Output dimensions after unsqueeze must be less than <= 6 | ------------------------------------------------------------------------------------ ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.9s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.10s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.284s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.1981s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2066s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2100s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2125s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2178s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2215s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2250s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2326s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.2361s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.2389s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.2425s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.2453s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.2487s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.2541s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.2571s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.2597s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.2625s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.2650s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.2694s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.2720s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.2748s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.2780s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.2808s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.2829s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.2850s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.2906s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.2936s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.2981s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.3025s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.3057s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.3084s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.3115s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.3147s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.3172s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.3207s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.3258s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.3293s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.3330s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.3364s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.3387s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.3421s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.3425s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.3439s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 23854573 bytes MEM: Free's : 26 free's of 23854573 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:50.917699137 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -4 is not in valid range [-2,1] Process Process-4: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -4 is not in valid range [-2,1]

[gw19] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw19] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw19] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1516'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1516', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -6
E assert -6 == 0
E + where -6 = <Process name='Process-5' pid=3549431 parent=3536831 stopped exitcode=-SIGABRT>.exitcode

test_tidl_unit.py:123: AssertionError[gw19] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.9s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.10s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.262s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2456s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2504s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2549s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2579s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.3101s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.3140s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.3197s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.3222s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3244s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3263s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3293s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3313s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.3338s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.3373s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.3395s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.3430s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.3455s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.3473s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.3494s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.3515s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.3532s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.3569s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.3589s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.3625s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.3655s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.3683s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.3706s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.3725s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.3748s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.3798s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.3835s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.3881s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.3928s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.3969s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.4006s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.4051s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.4130s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.4179s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.4212s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.4260s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.4300s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.4302s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.4317s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name
------------------------------Captured stderr call------------------------------
malloc(): corrupted top size

[gw23] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw23] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw23] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1111'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1111', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-5' pid=3549664 parent=3537186 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw23] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.12s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.12s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.365s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2419s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2714s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2751s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2798s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2857s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2892s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2997s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.3036s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3067s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3145s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3177s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3213s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.3246s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.3291s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.3317s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.3355s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.3388s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.3429s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.3475s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.3521s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.3555s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.3615s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.3648s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.3696s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.3737s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.3771s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.3802s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.3854s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.3906s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.3946s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.4006s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.4040s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.4068s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.4131s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.4168s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.4217s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.4268s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.4302s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.4338s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.4403s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.4446s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.4455s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.4484s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18659005 bytes MEM: Free's : 26 free's of 18659005 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:57.918579879 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 3 is not in valid range [-1,0] Process Process-5: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 3 is not in valid range [-1,0]

[gw20] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw20] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw20] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1080'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1080', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-6' pid=3549969 parent=3536928 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw20] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.20s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.20s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.324s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.1966s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2162s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2248s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2317s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2362s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2404s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2494s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2519s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.2565s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.2687s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.2726s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.2766s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.2800s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.2853s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.2896s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.2935s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.2968s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.2995s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.3032s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.3059s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.3101s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.3177s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.3222s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.3246s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.3288s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.3343s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.3382s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.3425s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.3458s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.3485s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.3538s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.3590s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.3619s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.3669s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.3698s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.3737s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.3789s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.3838s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.3875s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.3930s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.3983s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.3988s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.4007s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18656749 bytes MEM: Free's : 26 free's of 18656749 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:57:06.458242940 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 2 is not in valid range [-2,1] Process Process-6: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 2 is not in valid range [-2,1]

[gw1] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw1] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw1] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1859'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1859', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-7' pid=3550197 parent=3536268 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw1] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.14s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.15s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.315s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.1993s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2223s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2272s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2310s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2376s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2411s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2515s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2549s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.2580s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.2616s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.2654s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.2699s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.2735s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.2800s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.2839s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.2873s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.2923s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.2974s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.3014s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.3066s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.3100s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.3137s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.3171s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.3231s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.3270s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.3316s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.3348s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.3394s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.3438s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.3479s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.3538s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.3604s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.3647s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.3700s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.3739s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.3775s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.3833s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.3875s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.3913s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.3970s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.4019s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.4022s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.4037s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18681997 bytes MEM: Free's : 26 free's of 18681997 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:57:00.603357937 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -3 is not in valid range [-2,1] Process Process-7: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -3 is not in valid range [-2,1]

[gw15] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw15] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw15] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1016'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1016', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-6' pid=3550482 parent=3536629 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw15] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.15s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.17s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.417s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.4581s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.5015s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.5075s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.5159s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.5238s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.5308s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.5426s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.5484s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.5539s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.5602s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.5672s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.5709s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.5746s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.5811s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.5850s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.5887s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.5941s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.5988s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.6026s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.6092s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.6127s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.6174s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.6227s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.6264s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.6298s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.6357s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.6464s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.6524s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.6582s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.6628s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.6674s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.6731s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.6770s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.6807s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.6857s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.6903s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.6948s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.7013s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.7052s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.7105s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.7164s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.7168s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.7178s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18657121 bytes MEM: Free's : 26 free's of 18657121 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:57:08.124231721 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 3 is not in valid range [-2,1] Process Process-6: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 3 is not in valid range [-2,1]

[gw4] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw4] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw4] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1687'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1687', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-6' pid=3550583 parent=3536277 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw4] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.15s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.17s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.483s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.3561s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.3745s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.3803s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.3887s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.3962s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.4032s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.4187s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.4246s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.4306s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.4375s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.4426s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.4459s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.4508s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.5129s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.5268s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.5342s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.5411s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.5465s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.5529s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.5624s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.5710s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.5807s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.5886s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.5929s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.5984s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.6062s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.6128s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.6211s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.6267s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.6339s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.6433s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.6514s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.6580s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.6666s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.6739s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.6824s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.6948s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.7016s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.7084s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.7149s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.7233s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.7240s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.7262s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 23854573 bytes MEM: Free's : 26 free's of 23854573 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:56:58.229917508 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 4 is not in valid range [-2,1] Process Process-6: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 4 is not in valid range [-2,1]

[gw5] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw5] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw5] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1476'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1476', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -11
E assert -11 == 0
E + where -11 = <Process name='Process-5' pid=3551007 parent=3536280 stopped exitcode=-SIGSEGV>.exitcode

test_tidl_unit.py:123: AssertionError[gw5] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.9s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.10s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.256s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.1966s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2032s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2100s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2149s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2191s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2219s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2298s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2382s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.2425s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.2477s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.2506s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.2534s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.2566s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.2608s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.2642s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.2680s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.2712s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.2744s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.2783s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.2820s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.2857s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.3089s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.3127s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.3172s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.3206s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.3281s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.3319s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.3362s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.3398s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.3431s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.3472s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.3521s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.3558s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.3611s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.3644s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.3682s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.3730s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.3765s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.3801s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.3889s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.3938s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.3943s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.3954s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name ********** Frame Index 1069667847 Running inference - currFrameIdx > numFramesCalibration ********** 0.68155s: VX_ZONE_ERROR: [ownObjectDeInit:171] Is graph use failed, index: 0 0.68179s: VX_ZONE_ERROR: [ownObjectDeInit:177] Is node use failed, index: 0 0.68182s: VX_ZONE_ERROR: [ownObjectDeInit:183] Is kernel use failed, index: 0 0.68186s: VX_ZONE_ERROR: [ownObjectDeInit:184] kernel name: org.khronos.openvx.absdiff 0.68193s: VX_ZONE_ERROR: [ownObjectDeInit:196] Is user data object use failed, index: 0 0.68215s: VX_ZONE_ERROR: [ownObjectDeInit:240] Is tensor use failed, index: 0 0.68228s: VX_ZONE_ERROR: [ownObjectDeInit:282] Is error use failed, index: 0

[gw19] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw19] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw19] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1737'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1737', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -6
E assert -6 == 0
E + where -6 = <Process name='Process-6' pid=3551126 parent=3536831 stopped exitcode=-SIGABRT>.exitcode

test_tidl_unit.py:123: AssertionError[gw19] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.21s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.23s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.561s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.5787s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.6008s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.6199s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.6297s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.6461s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.6589s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.6781s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.6900s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.6997s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.7142s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.7253s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.7366s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.7528s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.7707s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.7861s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.7987s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.8102s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.8269s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.8396s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.8512s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.8697s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.8827s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.8925s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.9093s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.9197s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.9332s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.9476s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.9602s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.9730s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.9857s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.9965s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.10079s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.10220s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.10339s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.10442s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.10586s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.10728s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.10835s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.10983s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.11090s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.11198s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.11205s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.11235s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name
------------------------------Captured stderr call------------------------------
malloc(): invalid size (unsorted)

[gw16] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw16] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw16] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1263'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1263', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-5' pid=3551176 parent=3536664 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw16] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.18s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.19s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.334s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2727s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2935s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2972s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.3007s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.3073s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.3106s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.3189s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.3218s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3270s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3303s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3333s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3368s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.3396s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.3861s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.3966s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.4001s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.4029s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.4063s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.4113s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.4143s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.4190s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.4242s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.4277s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.4310s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.4336s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.4389s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.4428s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.4478s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.4513s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.4566s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.4610s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.4648s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.4686s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.4722s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.4760s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.4808s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.4849s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.4901s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.4945s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.4977s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.5033s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.5043s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.5058s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 20266093 bytes MEM: Free's : 26 free's of 20266093 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:57:03.554340351 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 2 is not in valid range [-2,1] Process Process-5: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 2 is not in valid range [-2,1]

[gw17] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw17] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw17] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1060'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1060', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-5' pid=3551629 parent=3536730 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw17] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.36s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.38s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.520s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.6192s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.6475s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.6563s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.6662s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.6785s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.6875s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.7035s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.7109s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.7173s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.7269s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.7355s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.7439s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.7523s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.8374s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.8585s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.8685s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.8778s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.8865s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.8949s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.9068s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.9151s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.9261s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.9385s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.9477s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.9554s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.9672s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.9742s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.9819s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.9921s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.10015s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.10129s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.10238s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.10330s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.10395s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.10493s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.10578s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.10667s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.10793s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.10886s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.10984s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.11130s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.11156s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.11182s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18657325 bytes MEM: Free's : 26 free's of 18657325 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:57:06.456585224 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -3 is not in valid range [-2,1] Process Process-5: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -3 is not in valid range [-2,1]

[gw22] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw22] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw22] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_651'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_651', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-6' pid=3552099 parent=3537059 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw22] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------ | Node | Node Name | Reason | ------------------------------------------------------------------------------------ | Unsqueeze | output | Output dimensions after unsqueeze must be less than <= 6 | ------------------------------------------------------------------------------------ ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.13s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.14s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.308s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.1988s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2114s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2218s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2246s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2295s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2329s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2365s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2445s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.2470s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.2502s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.2564s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.2604s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.2637s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.2694s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.2731s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.2764s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.2848s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.2878s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.2914s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.2972s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.3014s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.3053s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.3094s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.3125s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.3149s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.3190s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.3224s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.3256s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.3303s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.3349s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.3784s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.3850s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.3886s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.3909s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.3947s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.3996s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.4053s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.4119s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.4155s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.4197s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.4272s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.4278s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.4296s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 19629037 bytes MEM: Free's : 26 free's of 19629037 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:57:06.326024799 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 2 is not in valid range [-2,1] Process Process-6: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 2 is not in valid range [-2,1]

[gw6] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw6] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw6] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_641'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_641', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-6' pid=3552200 parent=3536285 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw6] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------ | Node | Node Name | Reason | ------------------------------------------------------------------------------------ | Unsqueeze | output | Output dimensions after unsqueeze must be less than <= 6 | ------------------------------------------------------------------------------------ ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.10s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.11s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.275s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.1952s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2101s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2137s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2171s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2232s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2279s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2307s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2382s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.2421s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.2454s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.2491s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.2526s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.2554s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.2610s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.2642s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.2684s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.2717s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.2754s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.2780s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.2813s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.2840s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.2868s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.2911s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.2939s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.2967s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.3022s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.3048s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.3071s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.3109s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.3145s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.3189s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.3219s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.3263s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.3287s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.3323s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.3370s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.3419s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.3453s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.3484s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.3520s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.3574s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.3579s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.3596s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 19629037 bytes MEM: Free's : 26 free's of 19629037 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:57:08.711918525 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 3 is not in valid range [-2,1] Process Process-6: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 3 is not in valid range [-2,1]

[gw1] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw1] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw1] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_891'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_891', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-8' pid=3553097 parent=3536268 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw1] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------ | Node | Node Name | Reason | ------------------------------------------------------------------------------------ | Unsqueeze | output | Output dimensions after unsqueeze must be less than <= 6 | ------------------------------------------------------------------------------------ ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.26s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.28s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.507s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.3294s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.3489s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.3624s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.3682s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.5850s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.5986s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.6065s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.6180s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.6242s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.6299s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.6367s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.6427s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.6483s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.8571s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.8771s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.8835s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.8901s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.8971s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.9016s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.9100s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.9153s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.9221s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.9268s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.9328s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.9382s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.9445s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.9486s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.9635s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.9675s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.9736s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.9799s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.9855s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.9897s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.9955s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.10007s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.10042s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.10105s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.10146s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.10195s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.10259s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.10318s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.10325s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.10354s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18954349 bytes MEM: Free's : 26 free's of 18954349 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:57:15.612904795 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -4 is not in valid range [-2,1] Process Process-8: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -4 is not in valid range [-2,1]

[gw19] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw19] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw19] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1088'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1088', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-7' pid=3553452 parent=3536831 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw19] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.15s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.18s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.408s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.3950s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.4097s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.4179s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.4244s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.4325s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.4403s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.4503s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.4574s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.4629s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.4705s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.4762s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.4821s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.4896s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.4976s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.5036s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.5101s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.5169s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.5251s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.5314s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.5389s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.5457s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.5545s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.5638s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.5725s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.5812s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.5943s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.6033s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.6121s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.6186s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.6246s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.6331s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.6415s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.6507s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.6576s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.6634s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.6707s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.6804s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.6863s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.6938s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.6993s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.7046s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.7049s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.7068s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18656749 bytes MEM: Free's : 26 free's of 18656749 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:57:21.057589120 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 3 is not in valid range [-1,0] Process Process-7: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 3 is not in valid range [-1,0]

[gw9] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw9] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw9] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1665'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1665', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-7' pid=3553610 parent=3536359 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw9] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.12s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.14s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.337s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2646s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2786s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2820s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2856s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2909s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2964s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.3027s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.3078s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3137s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3181s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3215s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3252s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.3288s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.3624s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.3678s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.3721s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.3762s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.3798s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.3853s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.3905s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.3949s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.3989s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.4056s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.4106s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.4160s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.4190s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.4223s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.4254s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.4298s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.4340s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.4406s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.4458s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.4495s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.4531s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.4568s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.4600s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.4655s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.4704s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.4766s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.4832s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.4881s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.4883s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.4898s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 19629037 bytes MEM: Free's : 26 free's of 19629037 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:57:16.050304516 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 4 is not in valid range [-1,0] Process Process-7: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 4 is not in valid range [-1,0]

[gw22] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw22] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw22] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_782'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_782', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-7' pid=3554051 parent=3537059 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw22] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------ | Node | Node Name | Reason | ------------------------------------------------------------------------------------ | Unsqueeze | output | Output dimensions after unsqueeze must be less than <= 6 | ------------------------------------------------------------------------------------ ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.8s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.9s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.240s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.1158s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.1227s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.1281s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.1297s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.1330s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.1350s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.1368s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.1415s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.1453s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.1481s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.1509s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.1529s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.1549s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.1576s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.1598s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.1620s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.1645s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.1662s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.1683s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.1706s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.1731s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.1747s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.1797s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.1815s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.1832s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.1855s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.1868s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.1888s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.1909s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.1927s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.1949s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.1972s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.1986s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.2001s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.2023s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.2045s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.2072s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.2103s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.2117s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.2144s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.2178s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.2181s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.2194s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18699757 bytes MEM: Free's : 26 free's of 18699757 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:57:27.621404833 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -3 is not in valid range [-2,1] Process Process-7: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -3 is not in valid range [-2,1]

[gw20] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw20] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw20] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1430'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1430', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -6
E assert -6 == 0
E + where -6 = <Process name='Process-7' pid=3554152 parent=3536928 stopped exitcode=-SIGABRT>.exitcode

test_tidl_unit.py:123: AssertionError[gw20] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.16s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.17s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.454s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.4039s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.4240s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.4403s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.4496s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.4607s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.4684s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.4852s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.4950s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.5021s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.5790s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.5908s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.5987s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.6097s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.6187s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.6240s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.6343s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.6400s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.6455s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.6552s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.6623s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.6681s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.6771s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.6825s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.6887s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.6994s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.7051s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.7103s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.7181s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.7229s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.7306s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.7399s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.7464s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.7517s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.7611s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.7662s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.7725s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.7818s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.7878s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.7964s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.8050s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.8117s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.8122s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.8132s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name ********** Frame Index 1062831271 Running inference - currFrameIdx > numFramesCalibration **********
------------------------------Captured stderr call------------------------------
malloc(): invalid size (unsorted)

[gw0] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw0] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw0] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1766'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1766', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -11
E assert -11 == 0
E + where -11 = <Process name='Process-7' pid=3554725 parent=3536265 stopped exitcode=-SIGSEGV>.exitcode

test_tidl_unit.py:123: AssertionError[gw0] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.12s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.13s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.280s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2885s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2959s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.3045s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.3081s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.3822s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.3868s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.4851s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.4947s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.4980s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.5039s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.5071s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.5107s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.5136s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.5196s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.5221s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.5261s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.5292s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.5317s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.5346s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.5383s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.5416s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.5443s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.5476s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.5499s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.5518s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.5540s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.5574s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.5608s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.5638s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.5664s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.5710s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.5757s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.5781s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.5826s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.5866s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.5900s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.5944s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.5973s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.6001s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.6035s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.6090s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.6098s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.6112s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name ********** Frame Index 1050838730 Running inference - currFrameIdx > numFramesCalibration ********** 0.32455s: VX_ZONE_ERROR: [ownObjectDeInit:171] Is graph use failed, index: 0 0.32461s: VX_ZONE_ERROR: [ownObjectDeInit:177] Is node use failed, index: 0 0.32463s: VX_ZONE_ERROR: [ownObjectDeInit:183] Is kernel use failed, index: 0 0.32465s: VX_ZONE_ERROR: [ownObjectDeInit:184] kernel name: org.khronos.openvx.absdiff 0.32468s: VX_ZONE_ERROR: [ownObjectDeInit:196] Is user data object use failed, index: 0 0.32479s: VX_ZONE_ERROR: [ownObjectDeInit:240] Is tensor use failed, index: 0 0.32485s: VX_ZONE_ERROR: [ownObjectDeInit:282] Is error use failed, index: 0

[gw14] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw14] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw14] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1321'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1321', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -11
E assert -11 == 0
E + where -11 = <Process name='Process-7' pid=3554749 parent=3536562 stopped exitcode=-SIGSEGV>.exitcode

test_tidl_unit.py:123: AssertionError[gw14] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.21s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.22s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.568s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.6470s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.6699s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.6875s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.6966s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.7090s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.7194s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.7365s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.7477s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.7556s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.7652s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.7726s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.7814s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.7893s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.8012s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.8092s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.8191s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.8282s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.8363s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.8447s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.8578s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.8680s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.8807s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.8897s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.9001s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.9099s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.9219s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.9323s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.9410s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.9502s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.9593s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.10083s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.10232s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.10317s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.10411s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.10505s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.10646s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.10751s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.10858s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.10949s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.11045s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.11186s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.11193s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.11224s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== 1.611423s: VX_ZONE_ERROR: [ownObjectDeInit:171] Is graph use failed, index: 0 1.611465s: VX_ZONE_ERROR: [ownObjectDeInit:177] Is node use failed, index: 0 1.611467s: VX_ZONE_ERROR: [ownObjectDeInit:183] Is kernel use failed, index: 0 1.611472s: VX_ZONE_ERROR: [ownObjectDeInit:184] kernel name: org.khronos.openvx.absdiff 1.611480s: VX_ZONE_ERROR: [ownObjectDeInit:196] Is user data object use failed, index: 0 1.611514s: VX_ZONE_ERROR: [ownObjectDeInit:240] Is tensor use failed, index: 0 1.611532s: VX_ZONE_ERROR: [ownObjectDeInit:282] Is error use failed, index: 0 1.659562s: VX_ZONE_ERROR: [ownPosixObjectDeInit:300] Deiniting event at index: 0 failed 1.659799s: VX_ZONE_ERROR: [ownPosixObjectDeInit:311] Deiniting queue at index: 41 failed 1.659814s: VX_ZONE_ERROR: [ownPosixObjectDeInit:333] Deiniting mutex at index: 10 failed MEM: Deinit ... !!! MEM: Alloc's: 25 alloc's of 24026861 bytes MEM: Free's : 25 free's of 24026861 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!

[gw12] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw12] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw12] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_704'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_704', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-7' pid=3554946 parent=3536493 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw12] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------ | Node | Node Name | Reason | ------------------------------------------------------------------------------------ | Unsqueeze | output | Output dimensions after unsqueeze must be less than <= 6 | ------------------------------------------------------------------------------------ ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.19s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.21s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.503s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.4091s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.4289s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.4407s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.4460s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.4558s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.4631s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.4708s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.4934s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.4960s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.5001s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.5020s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.5044s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.5077s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.5124s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.5152s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.5173s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.5236s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.5257s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.5283s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.5302s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.5335s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.5363s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.5387s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.5402s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.5428s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.5446s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.5477s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.5525s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.5543s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.5569s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.5690s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.5722s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.5763s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.5803s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.5829s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.5852s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.5909s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.5941s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.5967s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.6005s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.6032s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.6036s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.6047s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 23854573 bytes MEM: Free's : 26 free's of 23854573 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:57:20.224618655 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -3 is not in valid range [-2,1] Process Process-7: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -3 is not in valid range [-2,1]

[gw13] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw13] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw13] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1654'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1654', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-7' pid=3555290 parent=3536559 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw13] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.15s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.17s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.418s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2615s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2799s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2827s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2865s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2908s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2940s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.3032s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.3081s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3117s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3141s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3197s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3240s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.3287s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.3531s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.3582s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.3630s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.3670s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.3709s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.3736s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.3787s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.3824s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.3869s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.3905s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.3954s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.3991s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.4067s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.4110s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.4162s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.4204s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.4244s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.4283s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.4352s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.4399s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.4449s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.4486s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.4520s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.4568s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.4630s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.4674s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.4720s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.4774s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.4780s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.4796s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 19629037 bytes MEM: Free's : 26 free's of 19629037 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:57:25.730661898 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -4 is not in valid range [-2,1] Process Process-7: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -4 is not in valid range [-2,1]

[gw23] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw23] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw23] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1538'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1538', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -6
E assert -6 == 0
E + where -6 = <Process name='Process-7' pid=3555321 parent=3537186 stopped exitcode=-SIGABRT>.exitcode

test_tidl_unit.py:123: AssertionError[gw23] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.24s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.29s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.510s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.4845s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.5080s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.5158s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.5251s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.5370s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.5439s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.5597s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.5691s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.6139s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.6294s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.6364s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.6447s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.6562s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.6711s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.6786s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.6918s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.6994s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.7080s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.7190s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.7271s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.7335s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.7470s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.7566s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.7637s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.7735s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.7835s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.7910s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.8013s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.8795s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.8939s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.9089s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.9197s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.9281s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.9432s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.9519s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.9595s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.9729s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.9827s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.9916s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.10074s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.10173s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.10182s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.10209s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name
------------------------------Captured stderr call------------------------------
malloc(): corrupted top size

[gw8] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw8] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw8] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1849'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1849', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-8' pid=3555544 parent=3536324 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw8] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.14s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.15s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.390s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.3216s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.3356s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.3475s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.3536s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.3617s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.3674s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.3795s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.3831s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3864s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3910s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3951s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3993s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.4042s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.4103s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.4144s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.4201s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.4239s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.4282s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.4323s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.4358s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.4406s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.4441s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.4483s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.4516s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.4546s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.4592s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.4626s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.4673s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.4717s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.4749s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.4806s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.4872s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.4945s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.4981s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.5027s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.5066s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.5105s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.5153s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.5207s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.5256s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.5309s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.5312s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.5322s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18681997 bytes MEM: Free's : 26 free's of 18681997 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:57:31.213045902 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 4 is not in valid range [-1,0] Process Process-8: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 4 is not in valid range [-1,0]

[gw2] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw2] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw2] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1704'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1704', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-8' pid=3555613 parent=3536271 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw2] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.26s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.31s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.526s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.4168s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.4353s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.4413s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.4490s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.4583s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.4634s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.4732s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.4772s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.4832s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.4886s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.4930s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.4992s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.6862s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.7142s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.7209s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.7273s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.7333s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.7392s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.7438s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.7480s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.7531s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.7602s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.7648s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.7694s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.7740s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.8032s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.8093s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.8136s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.8183s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.8226s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.8295s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.8351s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.8399s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.8447s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.8501s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.8565s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.8613s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.8680s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.8731s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.8800s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.8882s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.8894s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.8921s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 23854573 bytes MEM: Free's : 26 free's of 23854573 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:57:25.980990700 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 2 is not in valid range [-2,1] Process Process-8: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 2 is not in valid range [-2,1]

[gw7] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw7] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw7] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1352'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1352', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -6
E assert -6 == 0
E + where -6 = <Process name='Process-7' pid=3555905 parent=3536289 stopped exitcode=-SIGABRT>.exitcode

test_tidl_unit.py:123: AssertionError[gw7] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.14s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.15s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.348s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2937s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.3104s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.3210s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.3249s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.3320s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.3350s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.3453s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.3491s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3549s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3586s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3620s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3649s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.3697s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.3746s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.3782s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.3835s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.3871s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.3925s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.3961s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.3998s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.4036s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.4093s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.4513s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.4595s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.4638s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.4693s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.4731s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.4770s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.4815s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.4852s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.4906s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.4954s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.4999s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.5026s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.5072s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.5111s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.5160s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.5200s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.5248s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.5309s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.5360s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.5367s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.5387s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name [TIDL Import] [PARSER] ERROR: Input ONNX tensor element type - 0. Only FLOAT, UINT8, INT32 and INT64 inputs supported for ONNX runtime -- [tidl_onnxRtImport_EP.cpp, 1861] [TIDL Import] [PARSER] ERROR: - Failed in function: TIDL_subgraphImport -- [tidl_onnxRtImport_EP.cpp, 1950] [TIDL Import] [PARSER] ERROR: - Failed in function: TIDL_computeInvokeFunc -- [tidl_onnxRtImport_EP.cpp, 2753]
------------------------------Captured stderr call------------------------------
2025-05-07 03:57:29.208406990 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running TIDL_0 node. Name:'TIDLExecutionProvider_TIDL_0_0' Status Message: TIDL Compute Invoke Failed. terminate called after throwing an instance of 'onnxruntime::OnnxRuntimeException' what(): /root/onnxruntime/onnxruntime/core/framework/bfc_arena.cc:92 onnxruntime::BFCArena::Chunk* onnxruntime::BFCArena::ChunkFromHandle(onnxruntime::BFCArena::ChunkHandle) h < chunks_.size() was false.

[gw10] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw10] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw10] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1040'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1040', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-6' pid=3556056 parent=3536424 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw10] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.15s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.16s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.330s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2708s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2849s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2906s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2945s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.3007s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.3047s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.3125s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.3168s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3199s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3230s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3268s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3298s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.3325s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.4037s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.4117s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.4182s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.4239s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.4288s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.4325s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.4395s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.4449s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.4491s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.4541s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.4590s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.4624s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.4702s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.4765s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.4798s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.4840s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.4917s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.4982s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.5045s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.5096s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.5144s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.5187s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.5219s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.5271s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.5318s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.5363s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.5437s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.5492s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.5498s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.5519s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18657325 bytes MEM: Free's : 26 free's of 18657325 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:57:30.142739991 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 1 is not in valid range [-1,0] Process Process-6: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 1 is not in valid range [-1,0]

[gw4] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw4] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw4] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1466'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1466', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -6
E assert -6 == 0
E + where -6 = <Process name='Process-8' pid=3556091 parent=3536277 stopped exitcode=-SIGABRT>.exitcode

test_tidl_unit.py:123: AssertionError[gw4] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.16s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.18s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.487s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.4151s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.4333s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.4392s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.4460s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.4538s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.4651s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.4772s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.4843s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.4910s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.4977s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.5041s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.5096s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.5158s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.5249s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.5312s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.5394s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.5464s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.5527s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.5696s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.5785s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.5859s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.5967s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.6049s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.6110s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.6202s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.6299s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.6382s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.6501s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.6589s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.6678s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.6820s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.6914s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.6995s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.7112s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.7188s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.7267s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.7392s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.7476s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.7554s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.7661s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.7754s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.7760s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.7783s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name ********** Frame Index 1069667847 Running inference - currFrameIdx > numFramesCalibration **********
------------------------------Captured stderr call------------------------------
malloc(): unaligned tcache chunk detected

[gw1] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw1] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw1] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1616'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1616', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: -6
E assert -6 == 0
E + where -6 = <Process name='Process-9' pid=3556183 parent=3536268 stopped exitcode=-SIGABRT>.exitcode

test_tidl_unit.py:123: AssertionError[gw1] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.15s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.16s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.449s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.3794s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.3943s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.4001s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.4051s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.4115s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.4173s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.4276s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.4327s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.4380s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.4435s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.4485s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.4534s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.4595s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.4678s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.4737s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.4791s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.4848s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.4920s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.4975s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.5036s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.5093s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.5148s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.5196s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.5263s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.5320s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.5390s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.5441s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.5500s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.5561s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.5625s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.5693s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.5740s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.5785s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.5853s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.5925s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.5994s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.6083s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.6151s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.6232s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.6334s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.6417s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.6424s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.6441s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name [TIDL Import] [PARSER] ERROR: Input ONNX tensor element type - -1107064492. Only FLOAT, UINT8, INT32 and INT64 inputs supported for ONNX runtime -- [tidl_onnxRtImport_EP.cpp, 1861] [TIDL Import] [PARSER] ERROR: - Failed in function: TIDL_subgraphImport -- [tidl_onnxRtImport_EP.cpp, 1950] [TIDL Import] [PARSER] ERROR: - Failed in function: TIDL_computeInvokeFunc -- [tidl_onnxRtImport_EP.cpp, 2753]
------------------------------Captured stderr call------------------------------
2025-05-07 03:57:26.210446358 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running TIDL_0 node. Name:'TIDLExecutionProvider_TIDL_0_0' Status Message: TIDL Compute Invoke Failed. terminate called after throwing an instance of 'onnxruntime::OnnxRuntimeException' what(): /root/onnxruntime/onnxruntime/core/framework/bfc_arena.cc:92 onnxruntime::BFCArena::Chunk* onnxruntime::BFCArena::ChunkFromHandle(onnxruntime::BFCArena::ChunkHandle) h < chunks_.size() was false.

[gw5] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw5] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw5] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1242'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1242', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-7' pid=3556434 parent=3536280 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw5] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.11s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.12s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.313s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2793s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2863s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2942s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.3011s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.3051s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.3081s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.3135s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.3192s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.3220s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.3256s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.3281s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.3317s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.3345s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.3389s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.3425s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.3464s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.3486s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.3521s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.3556s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.3589s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.3622s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.3659s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.3683s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.3712s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.3739s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.3774s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.3794s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.3824s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.3850s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.3882s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.3915s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.3947s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.3978s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.4002s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.4027s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.4058s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.4097s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.4131s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.4162s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.4207s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.4260s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.4263s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.4277s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 19374189 bytes MEM: Free's : 26 free's of 19374189 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:57:27.626307860 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 2 is not in valid range [-2,1] Process Process-7: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 2 is not in valid range [-2,1]

[gw18] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw18] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw18] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1161'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1161', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-7' pid=3556533 parent=3536765 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw18] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.13s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.15s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.314s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.2135s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.2301s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.2329s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.2389s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.2457s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.2491s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.2583s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.2631s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.2666s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.2695s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.2719s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.2756s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.2790s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.3523s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.3595s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.3648s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.3686s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.3722s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.3775s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.3821s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.3865s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.3939s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.3976s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.4028s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.4083s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.4121s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.4152s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.4199s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.4241s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.4272s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.4318s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.4389s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.4448s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.4489s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.4531s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.4574s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.4624s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.4670s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.4725s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.4763s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.4812s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.4820s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.4840s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18674797 bytes MEM: Free's : 26 free's of 18674797 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:57:34.936686181 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -3 is not in valid range [-1,0] Process Process-7: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis -3 is not in valid range [-1,0]

[gw17] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
[gw17] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3[gw17] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3

no_subprocess = False, tidl_offload = True, run_infer = False, operator_tests_root_fixture = 'tidl_unit_test_data/operator/'
test_name = 'Unsqueeze_1033'

@pytest.mark.parametrize(("test_name"), operator_tests_to_run)
def test_tidl_unit_operator(no_subprocess : bool, tidl_offload : bool, run_infer : bool, operator_tests_root_fixture : str, test_name : str):
'''
Pytest for tidl unit operator tests using the edgeai-benchmark framework
'''
testdir_parent = operator_tests_root_fixture
for i in range(len(operator_tests_to_run)):
if operator_tests_to_run[i] == test_name:
testdir_parent = operator_tests_parent_dir[i]
break

> perform_tidl_unit(no_subprocess=no_subprocess,
tidl_offload = tidl_offload,
run_infer = run_infer,
test_name = test_name,
test_suite = "operator",
testdir_parent = testdir_parent)

test_tidl_unit.py:70:
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
test_tidl_unit.py:89: in perform_tidl_unit
perform_tidl_unit_subprocess(tidl_offload = tidl_offload,
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

tidl_offload = True, run_infer = False, test_name = 'Unsqueeze_1033', test_suite = 'operator'
testdir_parent = 'tidl_unit_test_data/operator/Unsqueeze'

def perform_tidl_unit_subprocess(tidl_offload : bool, run_infer : bool, test_name : str, test_suite : str, testdir_parent : str):
'''
Perform an tidl unit test using a subprocess (in order to properly capture output for fatal errors)
Called by perform_tidl_unit
'''

kwargs = {"tidl_offload" : tidl_offload,
"run_infer" : run_infer,
"test_name" : test_name,
"test_suite" : test_suite,
"testdir_parent" : testdir_parent}

p = Process(target=perform_tidl_unit_oneprocess, kwargs=kwargs)
p.start()


# Note: This timeout parameter must be lower than the pytest-timeout parameter
# passed to the pytest command (on the command line or in pytest.ini)
p.join(timeout=600)
if p.is_alive():
p.terminate()

# Cleanup leftover files
for f in glob.glob("/dev/shm/vashm_buff_*"):
os.remove(f)

> assert p.exitcode == 0, f"Received nonzero exit code: {p.exitcode}"
E AssertionError: Received nonzero exit code: 1
E assert 1 == 0
E + where 1 = <Process name='Process-7' pid=3557074 parent=3536730 stopped exitcode=1>.exitcode

test_tidl_unit.py:123: AssertionError[gw17] linux -- Python 3.10.16 /home/tidl/.pyenv/versions/benchmark/bin/python3
------------------------------Captured stdout call------------------------------
========================= [Model Compilation Started] ========================= Model compilation will perform the following stages: 1. Parsing 2. Graph Optimization 3. Quantization & Calibration 4. Memory Planning ============================== [Version Summary] ============================== ------------------------------------------------------------------------------- | TIDL Tools Version | 11_00_00_00 | ------------------------------------------------------------------------------- | C7x Firmware Version | 11_00_00_00 | ------------------------------------------------------------------------------- | Runtime Version | 1.15.0 | ------------------------------------------------------------------------------- | Model Opset Version | 19 | ------------------------------------------------------------------------------- ============================== [Parsing Started] ============================== [TIDL Import] [PARSER] WARNING: Network not identified as Object Detection network : (1) Ignore if network is not Object Detection network (2) If network is Object Detection network, please specify "model_type":"OD" as part of OSRT compilation options ------------------------- Subgraph Information Summary ------------------------- ------------------------------------------------------------------------------- | Core | No. of Nodes | Number of Subgraphs | ------------------------------------------------------------------------------- | C7x | 1 | 1 | | CPU | 1 | x | ------------------------------------------------------------------------------- ------------------------------------------------------------------------------------------------------- | Node | Node Name | Reason | ------------------------------------------------------------------------------------------------------- | Unsqueeze | output | Layer 1 - op type Unsqueeze, Unknown input dimension, not supported by TIDL | ------------------------------------------------------------------------------------------------------- ============================= [Parsing Completed] ============================= ==================== [Optimization for subgraph_0 Started] ==================== ----------------------------- Optimization Summary ----------------------------- ---------------------------------------------------------------------------- | Layer | Nodes before optimization | Nodes after optimization | ---------------------------------------------------------------------------- | TIDL_EltWiseLayer | 1 | 0 | ---------------------------------------------------------------------------- Total nodes in subgraph: 4 =================== [Optimization for subgraph_0 Completed] =================== The soft limit is 10240 The hard limit is 10240 MEM: Init ... !!! MEM: Init ... Done !!! 0.0s: VX_ZONE_INFO: Globally Enabled VX_ZONE_INFO 0.26s: VX_ZONE_INFO: Globally Enabled VX_ZONE_ERROR 0.28s: VX_ZONE_INFO: Globally Enabled VX_ZONE_WARNING 0.544s: VX_ZONE_INFO: [ownAddTargetKernelInternal:189] registered kernel vx_tutorial_graph.phase_rgb on target DSP-1 0.5738s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-0 0.5979s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-1 0.6153s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-2 0.6284s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MPU-3 0.6422s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1 0.6540s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_2 0.6753s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_3 0.6872s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_4 0.6983s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_5 0.7089s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_6 0.7211s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_7 0.7313s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP_C7-1_PRI_8 0.7455s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSP-1 0.7603s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-0 0.7707s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_NF 0.7840s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_LDC1 0.7940s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC1 0.10012s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_MSC2 0.10371s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_VISS1 0.10495s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE1 0.10566s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE2 0.10712s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE3 0.10781s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE4 0.10864s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE5 0.10979s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE6 0.11044s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE7 0.11104s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CAPTURE8 0.11214s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY1 0.11278s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DISPLAY2 0.11365s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX 0.11495s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target CSITX2 0.11567s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M1 0.11658s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M2 0.11784s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M3 0.11876s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DSS_M2M4 0.11969s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target VPAC_FC 0.12109s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU2-1 0.12224s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_SDE 0.12321s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target DMPAC_DOF 0.12464s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-0 0.12568s: VX_ZONE_INFO: [tivxPlatformCreateTargetId:169] Added target MCU3-1 0.12575s: VX_ZONE_INFO: [tivxInit:152] Initialization Done !!! 0.12610s: VX_ZONE_INFO: Globally Disabled VX_ZONE_INFO ============= [Quantization & Calibration for subgraph_0 Started] ============= Warning : Couldn't find corresponding ioBuf tensor for onnx tensor with matching name -------- Running Calibration in Float Mode to Collect Tensor Statistics -------- [=============================================================================] 100 % ------------------ Fixed-point Calibration Iteration [1 / 1]: ------------------ [=============================================================================] 100 % ==================== [Quantization & Calibration Completed] ==================== [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. Rerunning network compiler... [TIDL Import] WARNING: Couldn't open perfSimTool file: /home/tidl/pranav/edgeai-benchmark/tools/tidl_tools_package/AM68A/tidl_tools/ti_cnnperfsim.out. Skipping Performance Simulation. [TIDL Import] [PARSER] WARNING: ******************************************************************** * Network compiler returned with error or didn't executed * * This model can only be used on PC/Host emulation mode * * It is not expected to work on target/EVM * ******************************************************************** ======================== Subgraph Compiled Successfully ======================== MEM: Deinit ... !!! MEM: Alloc's: 26 alloc's of 18657121 bytes MEM: Free's : 26 free's of 18657121 bytes MEM: Open's : 0 allocs of 0 bytes MEM: Deinit ... Done !!!
------------------------------Captured stderr call------------------------------
2025-05-07 03:57:41.380259924 [E:onnxruntime:, sequential_executor.cc:514 ExecuteKernel] Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 2 is not in valid range [-2,1] Process Process-7: Traceback (most recent call last): File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 314, in _bootstrap self.run() File "/home/tidl/.pyenv/versions/3.10.16/lib/python3.10/multiprocessing/process.py", line 108, in run self._target(*self._args, **self._kwargs) File "/home/tidl/pranav/edgeai-benchmark/tests/tidl_unit/test_tidl_unit.py", line 206, in perform_tidl_unit_oneprocess results_list = onnxruntime_wrapper.run_import(tidl_unit_dataset[0]) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 58, in run_import output = self._run(input_data, output_keys) File "/home/tidl/pranav/edgeai-benchmark/edgeai_benchmark/core/onnxrt_runtime.py", line 91, in _run outputs = self.interpreter.run(output_keys, input_data) File "/home/tidl/.pyenv/versions/benchmark/lib/python3.10/site-packages/onnxruntime/capi/onnxruntime_inference_collection.py", line 217, in run return self._sess.run(output_names, input_feed, run_options) onnxruntime.capi.onnxruntime_pybind11_state.RuntimeException: [ONNXRuntimeError] : 6 : RUNTIME_EXCEPTION : Non-zero status code returned while running Unsqueeze node. Name:'' Status Message: /root/onnxruntime/onnxruntime/core/providers/common.h:23 int64_t onnxruntime::HandleNegativeAxis(int64_t, int64_t) axis >= -tensor_rank && axis <= tensor_rank - 1 was false. axis 2 is not in valid range [-2,1]